**Epidemiology of Colorectal Cancer — Incidence, Lifetime Risk Factors Statistics and Temporal Trends**

Camille Thélin and Sanjay Sikka

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61945

#### **Abstract**

Colorectal cancer is a major cause of morbidity and mortality in the entire world. Among cancers that affect both men and women, it accounts for >8% of all cancer incidence, making it the third most common cancer worldwide (behind lung and breast cancer). There were an estimated 14.1 million cancer cases around the world in 2012-last data available; 7.4 million were in men and 6.7 million in women. Of that, nearly 1.4 million new cases were from colorectal cancer. And, it has consistently been shown that the developed world carries the majority of the burden (Australia, New Zealand, Canada, the United States and parts of Western Europe), likely due to similarity in lifestyles and diets.

**Keywords:** Colon cancer epidemiology, colorectal cancer, SEER

#### **1. Introduction**

Colorectal cancer is a major cause of morbidity and mortality in the entire world. It has consistently been shown that the developed world carries the majority of the burden — this includes Australia, New Zealand, Canada, the United States and parts of Western Europe likely due to similarity in lifestyles and diets. [9, 12]

Among cancers that affect both men and women, colorectal cancer accounts for >8% of all cancer incidence, making it the third most common cancer worldwide, behind lung and breast cancer (Table 1). [1]

© 2015 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


\*Excludes basal cell and squamous cell skin cancers and *in situ* carcinoma except urinary bladder. Source: GLOBOCAN 2012 v1.1, "Cancer Incidence and Mortality Worldwide"

**Table 1.** Cancer Incidence Worldwide

There were an estimated 14.1 million cancer cases around the world in 2012. [1] Of those cancers, 7.4 million were in men, while 6.7 million were in women. [1] Nearly, 1.4 million of those new cancer cases were from colorectal cancer. [1]

In the United States, the breakdown between genders is similar. Colorectal cancer is the third most common cancer in both women and men (after breast and prostate cancer, respectively, and lung cancer). Among both gender groups, it is the second leading cause of cancer deaths (behind lung cancer), with peak incidence being in the seventh decade of life. [24] In 2015, it is estimated that there will be 848,200 new cases of cancer among men and 810,000 among women in 2015 (Table 2). [2] Of those new cancer cases, 8% will comprise of colon and rectal cancer, with an estimated 69,090 in men and 63,610 in females. [2]


Epidemiology of Colorectal Cancer — Incidence, Lifetime Risk Factors Statistics and Temporal Trends http://dx.doi.org/10.5772/61945 63


\*Excludes basal cell and squamous cell skin cancers and *in situ* carcinoma except urinary bladder.

Source: American Cancer Society, ―Cancer Facts and Figures 2015.‖ Projected cases are based on incidence data during 1995-2011 from 49 states and the District of Columbia, as reported by the North American Association of

Central Cancer Registries (NAACCR).

**Cancer New cases diagnosed in**

1 Lung 1,825 13.0 2 Breast 1,677 11.9 **3 Colorectal 1,361 9.7**

1 Lung 1,242 16.7 2 Prostate 1,112 15.0 **3 Colorectal 746 10.0**

1 Breast 1,677 25.2 2 **Colorectal 614 9.2** 3 **Lung 583 8.8**

\*Excludes basal cell and squamous cell skin cancers and *in situ* carcinoma except urinary bladder. Source: GLOBOCAN

There were an estimated 14.1 million cancer cases around the world in 2012. [1] Of those cancers, 7.4 million were in men, while 6.7 million were in women. [1] Nearly, 1.4 million of

In the United States, the breakdown between genders is similar. Colorectal cancer is the third most common cancer in both women and men (after breast and prostate cancer, respectively, and lung cancer). Among both gender groups, it is the second leading cause of cancer deaths (behind lung cancer), with peak incidence being in the seventh decade of life. [24] In 2015, it is estimated that there will be 848,200 new cases of cancer among men and 810,000 among women in 2015 (Table 2). [2] Of those new cancer cases, 8% will comprise of colon and rectal

**Male Female**

Prostate 26% 29% Breast Lung & bronchus 14% 13% Lung & bronchus **Colon & rectum 8% (69,090) 8% (63,610) Colon & rectum** Urinary bladder 7% 7% Uterine corpus

**Worldwide**

62 Screening for Colorectal Cancer with Colonoscopy

**Men**

**Women**

2012 v1.1, "Cancer Incidence and Mortality Worldwide"

those new cancer cases were from colorectal cancer. [1]

cancer, with an estimated 69,090 in men and 63,610 in females. [2]

**Table 1.** Cancer Incidence Worldwide

**Men 848,200** **2012 (1,000s)**

**Percent of all cancers\***

**Women 810,000**

Note: Estimates should not be compared with those from previous years.

**Table 2.** Estimated New Cancer Cases\* in the U.S. in 2015

#### **2. Clinical presentation of colorectal cancer**

The importance of screening is crucial as most early-stage colorectal cancer does not typically have symptoms. In fact, colorectal cancer may be quiescently growing for as long as 5 years before symptoms appear.

#### **2.1. Signs and symptoms**

Symptoms can be specific, such as abdominal discomfort and alarming changes in bowel movements (i.e., hematochezia, diarrhea, or obstruction). More often than not, however, symptoms are usually nonspecific, such as fatigue, weight loss, and/or changes in digestion. As such, even those with some type of symptoms have been misdiagnosed with other benign conditions. These benign conditions include examples such as diverticular disease, inflam‐ matory bowel syndrome, or hemorrhoids. [4]

The major biochemical sign is that of new onset anemia. In fact, in those older than 40 years old, a new onset anemia — specifically hypochromic and microcytic — should prompt evaluation for colorectal cancer.

#### **2.2. Right-sided colon cancers**

Symptoms depend somewhat on the site of the tumor. In general, **right-sided colon cancers** are usually detected at an advanced stage with severe symptoms. In general, the right-sided colon cancers are commonly larger, producing vague abdominal discomfort and sometimes a palpable mass. [4, 5] Obstruction is rarely a presenting symptom, as the diameter of the right colon is larger than the left colon. [4] If the tumor involves the cecum, however, it could block the ileocecal valve causing small bowel obstruction.

Those with right-sided colon cancers are significantly older and are predominantly women (46% women versus 38% men). [6] Because of higher rates of comorbidities, survival is worse in those with right-sided carcinomas.

#### **2.3. Left-sided colon cancers and rectal cancers**

In comparison, **left-sided colon cancers and rectal cancers** tend to arise in younger, male populations with high-incidence risk. [7, 8] Cancers involving this portion of the bowel produce symptoms that range from obstruction to tenesmus, to alternating constipation and diarrhea with pencil-thin stools. [4] Often, there is blood witnessed either in the stool or coating the stool, in comparison to the right-sided colon cancers. Similarly, rectal cancers can cause obstruction and similar types of bowel movement changes as the left-sided carcinomas.

#### **3. Risk factors of colorectal cancer**

There are both modifiable and nonmodifiable risk factors associated with the incidence of colorectal cancer (Table 3).


**Table 3.** Factors Associated with Higher Risk of Colon and Rectal Cancer

**Modifiable risk factors** include diet, physical activity, weight, cigarette-smoking, and alcohol intake. [9] Other modifiable risk factors include low calcium content, low selenium content, and very low salt intake. [10] Occupational hazards, such as asbestos-exposure, have been linked to increased risk of colorectal cancer when compared to the rest of the general popula‐ tion. [10]

Socioeconomic factors, along with access to (and use of) health care services, are also important contributing risk factors. In fact, there is a disproportionately high incidence of colorectal cancers in low socioeconomic status populations. [11]

**Nonmodifiable risk factors** associated with higher risk of colorectal cancer include increasing age, personal history of adenomatous polyps, personal history of inflammatory bowel disease, genetic inheritance, race/ethnicity, and gender. [9] Unlike modifiable risk factors that could theoretically have been avoided, these risk factors are not considered part of the "environ‐ mental nature" of this disease. Thus, they are not controllable. They do, however, play an important role in screening and identifying susceptible patients.

#### **3.1. Modifiable risk factors: Diet**

**2.2. Right-sided colon cancers**

64 Screening for Colorectal Cancer with Colonoscopy

the ileocecal valve causing small bowel obstruction.

**2.3. Left-sided colon cancers and rectal cancers**

**3. Risk factors of colorectal cancer**

Social behaviors (i.e., alcohol and cigarette smoking)

**Table 3.** Factors Associated with Higher Risk of Colon and Rectal Cancer

colorectal cancer (Table 3).

Diet

Physical activity Body weight

tion. [10]

in those with right-sided carcinomas.

Symptoms depend somewhat on the site of the tumor. In general, **right-sided colon cancers** are usually detected at an advanced stage with severe symptoms. In general, the right-sided colon cancers are commonly larger, producing vague abdominal discomfort and sometimes a palpable mass. [4, 5] Obstruction is rarely a presenting symptom, as the diameter of the right colon is larger than the left colon. [4] If the tumor involves the cecum, however, it could block

Those with right-sided colon cancers are significantly older and are predominantly women (46% women versus 38% men). [6] Because of higher rates of comorbidities, survival is worse

In comparison, **left-sided colon cancers and rectal cancers** tend to arise in younger, male populations with high-incidence risk. [7, 8] Cancers involving this portion of the bowel produce symptoms that range from obstruction to tenesmus, to alternating constipation and diarrhea with pencil-thin stools. [4] Often, there is blood witnessed either in the stool or coating the stool, in comparison to the right-sided colon cancers. Similarly, rectal cancers can cause obstruction and similar types of bowel movement changes as the left-sided carcinomas.

There are both modifiable and nonmodifiable risk factors associated with the incidence of

**Modifiable risk factors** include diet, physical activity, weight, cigarette-smoking, and alcohol intake. [9] Other modifiable risk factors include low calcium content, low selenium content, and very low salt intake. [10] Occupational hazards, such as asbestos-exposure, have been linked to increased risk of colorectal cancer when compared to the rest of the general popula‐

Age (≥50 years old)

Personal history of adenomatous colonic polyps

Personal history of inflammatory bowel disease (IBD)

Family history of colorectal cancer Hereditary polyposis conditions

**Modifiable Risk Factors Nonmodifiable Risk Factors**

Diets associated with high incidence of colorectal cancer include diets with high consumption of red or processed meat, diets high in fat, beer-drinking, diets low in calcium intake, and diets low in whole-grain fiber, fruits and vegetables. [9] This represents a typical "Western diet."

**On average, 40–45% of Western diets have total caloric intake made up from fatty foods (including meat products), while fat only accounts for about 10–15% of dietary makeup in lower-risk populations — China, India, and parts of Africa and South America.** Conse‐ quently, it has been shown that the developed world carries the majority of the burden (Australia, New Zealand, Canada, the United States and parts of Western Europe), [9, 12] likely due to similarity in lifestyles and diets.

The hypothesis behind dietary fat as a risk factor is that the fat enhances hepatic cholesterol and bile acid synthesis resulting in increased sterols in the colon. [4] Those sterols are then converted into secondary bile acids, cholesterol metabolites, and potentially toxic metabolic compounds. [4, 13]

While the exact pathogenesis remains unknown, what is known is that these sterols and bile acid metabolites cause damage to colonic mucosa, thus enhancing proliferative activity which could lead to dysplasia. [4, 13] This has been demonstrated in animal models, where animals fed polyunsaturated and saturated fats have higher numbers of adenocarcinoma than those on a low-fat diet. [4] This has also been shown in human population studies where those with colorectal cancer tend to have higher fecal bile acid levels, [4] while a recent meta-analysis has shown that consumption of red meat and processed meat is positively associated with risk of both colon — particularly the descending and sigmoid colon — and rectal cancer. [14]

The "Western diet" also comprises of lower amounts of fiber intake. Multiple epidemiology studies have shown a geographical difference of lower colorectal cancer incidence rates in places with higher fiber intake. [9] It is even postulated that due to the ability of fiber to change the colonic pH, carcinogenesis may be impeded. [4, 9]

Dietary fiber also increases fecal bulk, thus diluting the aforementioned carcinogenic com‐ pounds and reducing transit time and mucosal contact. In fact, fiber has been found to decrease the concentration of sterol and bile acid metabolites that could be implicated in creating carcinogenic compounds. [4] Again, this has been demonstrated in animal models, where increased fiber intake led to decreased concentration of specific bacterial metabolic enzymes that could be implicated in creating carcinogenic compounds. [4] Unfortunately, for all its experimentally demonstrative protective roles, increased fiber supplementation has been unable to prevent adenoma recurrence in several randomized-controlled trials.

#### **3.2. Modifiable risk factors: Physical activity and body weight**

Other modifiable risk factors are physical inactivity and excess body weight. Decreased gut motility, increased insulin resistance, lower metabolic rates, and increased circulating estro‐ gens are all mechanisms implicated in the higher risk of colorectal cancer associated with this modifiable risk factor. [9, 10]

#### **3.3. Modifiable risk factors: Social behaviors**

Associated with a higher risk is regular consumption of cigarettes and alcohol. [10] Carcino‐ genic metabolites found in both tobacco and alcohol are considered promoters of tumor growth, based on experimental studies in animals. [15]

Cigarette-smoking has been attributed to 12% of colorectal cancer deaths, while alcohol consumption has been linked with early onset colorectal cancers, specifically tumors in the distal colon. [9, 16, 17] There is information showing that there is higher risk in active smokers for development of rectal cancer.[9, 18]

#### **3.4. Nonmodifiable risk factors: Age**

Increasing age carries a higher likelihood of colorectal cancer, specifically after the age of 40. [2]

Cancer incidence rises progressively after the age of 40 in the general population, with 90% of colorectal cancers occurring in those aged 50 years and older. [2] In fact, a 50-year old has 5% chance of developing cancer and 2.5% chance of dying from this cancer after the age of 80 years. [2, 9]

As such, the US Preventative Task Force (USPSTF) has defined "average risk" as those aged 50 years or more with no personal history of colorectal cancer or adenomas, no inflammatory bowel disease, and with negative family history. [19] Put in other terms, the incidence rate is more than 50 times higher in those 60–79 years old than in those less than 40 years old.

In contrast, those with "increased risk" include those with a personal history of colorectal cancer, personal history of colonic adenomas, family history of sporadic colorectal cancer, as well as family history of sporadic adenoma. [4, 9]

Finally, those with "high risk" include those with hereditary nonpolyposis colorectal cancer (Lynch syndrome), polyposis syndromes, and inflammatory bowel diseases (IBD). [4] *See below for a discussion on hereditary polyposis conditions and IBD.*

#### **3.5. Nonmodifiable risk factors: Personal history of colonic adenomatous polyps**

Carrying a personal history of adenomatous polyps has an increased risk of developing colorectal cancer, in comparison to those with no history of adenomas. In recent literature, it was reported that 95% of sporadic colorectal cancers developed from such adenomas, usually after a protracted period, which has been estimated anywhere from 5 to 10 years. [4, 9] However, while nearly all colorectal cancer arise from adenomas, only a small minority of these dysplastic polyps actually progress to cancer (5% or less). [4]

#### **3.6. Nonmodifiable risk factors: Family history of colonic adenomatous polyps or colorectal cancer**

The majority of cases occur in those with family history of either colorectal cancer or adenom‐ atous cancer. In fact, there is a two- to three-fold increased risk of sporadic cancer in those with first-degree relatives. This means that up to 20% of those with colorectal cancer have family members affected by this disease. [4, 9] This risk becomes even higher when there are two or more relatives involved and when those family members are affected by the disease at an age younger than 60.

#### **3.7. Nonmodifiable risk factors: Hereditary polyposis conditions**

Those with recognized inherited polyposis syndromes carry an even higher risk. Recent literature estimates that about 5–10% of sporadic colorectal cancers are the outcome of inherited conditions, such as the familial adenomatous polyposis (FAP) and hereditary nonpolyposis colorectal cancer (HPNCC). [4, 9]

HPNCC (also called Lynch syndrome) is thought to comprise of about 1–6% of all colorectal cancers. It carries a lifetime risk of cancer as high as 70–80%. [4, 9] FAP and its variants account for less than 1% of all colorectal cancer cases, but almost all those diagnosed with this disorder will develop cancer if the colon is not removed by the age of 40. [4]

Other hereditary conditions that are associated with sporadic colorectal cancers include Gardner's syndrome (high-risk), Turcot's syndrome (high-risk), and Peutz-Jeghers syndrome (low-to-moderate risk). [4] Appropriate screening recommendations are made for this population subtype, which will not be discussed here.

#### **3.8. Nonmodifiable risk factors: Personal history of Inflammatory Bowel Disease (IBD)**

Those with IBD — ulcerative colitis and Crohn's disease — also carry an increased risk of developing colorectal cancer. It has been estimated that the relative risk of colorectal cancer in patients with IBD ranges from 4- to 20-fold. [4, 9] Thus, appropriate screening recommenda‐ tions are made for this population subtype, which will not be discussed here.

#### **4. Statistics**

increased fiber intake led to decreased concentration of specific bacterial metabolic enzymes that could be implicated in creating carcinogenic compounds. [4] Unfortunately, for all its experimentally demonstrative protective roles, increased fiber supplementation has been

Other modifiable risk factors are physical inactivity and excess body weight. Decreased gut motility, increased insulin resistance, lower metabolic rates, and increased circulating estro‐ gens are all mechanisms implicated in the higher risk of colorectal cancer associated with this

Associated with a higher risk is regular consumption of cigarettes and alcohol. [10] Carcino‐ genic metabolites found in both tobacco and alcohol are considered promoters of tumor

Cigarette-smoking has been attributed to 12% of colorectal cancer deaths, while alcohol consumption has been linked with early onset colorectal cancers, specifically tumors in the distal colon. [9, 16, 17] There is information showing that there is higher risk in active smokers

Increasing age carries a higher likelihood of colorectal cancer, specifically after the age of 40. [2] Cancer incidence rises progressively after the age of 40 in the general population, with 90% of colorectal cancers occurring in those aged 50 years and older. [2] In fact, a 50-year old has 5% chance of developing cancer and 2.5% chance of dying from this cancer after the

As such, the US Preventative Task Force (USPSTF) has defined "average risk" as those aged 50 years or more with no personal history of colorectal cancer or adenomas, no inflammatory bowel disease, and with negative family history. [19] Put in other terms, the incidence rate is more than 50 times higher in those 60–79 years old than in those less than 40 years old.

In contrast, those with "increased risk" include those with a personal history of colorectal cancer, personal history of colonic adenomas, family history of sporadic colorectal cancer, as

Finally, those with "high risk" include those with hereditary nonpolyposis colorectal cancer (Lynch syndrome), polyposis syndromes, and inflammatory bowel diseases (IBD). [4] *See below*

Carrying a personal history of adenomatous polyps has an increased risk of developing colorectal cancer, in comparison to those with no history of adenomas. In recent literature, it

**3.5. Nonmodifiable risk factors: Personal history of colonic adenomatous polyps**

unable to prevent adenoma recurrence in several randomized-controlled trials.

**3.2. Modifiable risk factors: Physical activity and body weight**

modifiable risk factor. [9, 10]

66 Screening for Colorectal Cancer with Colonoscopy

**3.3. Modifiable risk factors: Social behaviors**

for development of rectal cancer.[9, 18]

**3.4. Nonmodifiable risk factors: Age**

well as family history of sporadic adenoma. [4, 9]

*for a discussion on hereditary polyposis conditions and IBD.*

age of 80 years. [2, 9]

growth, based on experimental studies in animals. [15]

#### **4.1. Methods**

The following statistical data were obtained from the Surveillance, Epidemiology and End Results (SEER) Program of the National Cancer Institute (NCI), specifically from the data previously published in the *SEER Cancer Statistic Review (CSR) 1975*–*2012*, which was released in April 23, 2015. The NCI funds for the program through Centers for Disease Control and Presentation (CDC), National Program of Cancer Registries, and involved states' contribu‐ tions.

The SEER program was conceptualized in 1973, with a mission to report the "most recent cancer incidence, mortality, survival, prevalence, and lifetime risks statistics. It originally only represented about 10% of the US population. **Statistical Temporal Trends Statistical Temporal Trends**

Since then, it has expanded to include the following population-based cancer registries: Alaska Native Tumor Registry, Arizona Indians, Cherokee Nation, Connecticut, Detroit, Georgia Center for Cancer Statistics (Atlanta, Greater Georgia, Rural Georgia), Greater Bay Area Cancer Registry (San Francisco-Oakland, San Jose-Monterey), Greater California, Hawaii, Iowa, Kentucky, Los Angeles, Louisiana, New Jersey, New Mexico, Seattle-Puget Sound, and Utah. This translates to approximately 26% of African Americans, 41% of Hispanics, 43% of Amer‐ ican Indians and Alaska Natives, 54% of Asians, and 71% of Hawaiian/Pacific Islanders. It is published annually, with 2012 being the most recent year for which data are available. The following statistical data was obtained from the Surveillance, Epidemiology and End Results (SEER) Program of the National Cancer Institute (NCI), specifically from data previously published in the *SEER Cancer Statistic Review (CSR) 1975-2012*, which was released April 23, 2015. The NCI funds for the program through Centers for Disease Control and Presentation (CDC), National Program of Cancer Registries, and involved states' contributions. The SEER program was conceptualized in 1973, with a mission to report the ―most recent cancer incidence, mortality, survival, prevalence, and lifetime risks statistics. It originally only represented ~10% of the US population. Since then, it has expanded to include the following population-based cancer registries: Alaska Native Tumor Registry, Arizona Indians, Cherokee Nation, Connecticut, Detroid, Georgia Center for Cancer Statistics (Atlanta, Greater Georgia, Rural Georgia), Greater Bay Area Cancer

#### **4.2. Temporal trends in the united states** Registry (San Francisco-Oakland, San Jose-Monterey), Greater California, Hawaii, Iowa, Kentucky, Los Angeles, Louisiana, New Jersey, New Mexico, Seattle-Puget Sound, and Utah.

**How common is this cancer?** It is estimated that there will be 132,700 new colorectal cancer cases in 2015. [21] This comprises 8% of all new cancer cases (Figure 1). [21] Of those new cancer cases, there will be an estimated 49,700 deaths. [21] This comprises 8.4% of all cancer deaths (Table 4). Indians and Alaska Natives, 54% of Asians, and 71% of Hawaiian/Pacific Islanders. It is published annually, with 2012 being the most recent year for which data is available. **How common is this cancer?** It is estimated that there will be 132,700 new cases in 2015 (Table 4). This comprises 8% of all new cancer cases. Of those new cases, it is estimated that there will be 49,700 deaths. This comprises 8.4% of all cancer deaths in the United States (Figure 1).

This translates to approximately 26% of African Americans, 41% of Hispanics, 43% of American

**Figure 1.** Colon and rectum cancer in the U.S.

**Who gets this cancer?** Colorectal cancer is more common in men than in women. In 2014, there were a total of 135,260 people diagnosed with colorectal cancer: 70,099 men versus 65,161 women. [22] Based on SEER 18, this means that 48.9 per 100,000 persons new cases were male, while 37.1 per 100,000 persons were female. [20]


**Table 4.** Comparison of Common Cancers

previously published in the *SEER Cancer Statistic Review (CSR) 1975*–*2012*, which was released in April 23, 2015. The NCI funds for the program through Centers for Disease Control and Presentation (CDC), National Program of Cancer Registries, and involved states' contribu‐

The SEER program was conceptualized in 1973, with a mission to report the "most recent cancer incidence, mortality, survival, prevalence, and lifetime risks statistics. It originally only

Since then, it has expanded to include the following population-based cancer registries: Alaska Native Tumor Registry, Arizona Indians, Cherokee Nation, Connecticut, Detroit, Georgia Center for Cancer Statistics (Atlanta, Greater Georgia, Rural Georgia), Greater Bay Area Cancer Registry (San Francisco-Oakland, San Jose-Monterey), Greater California, Hawaii, Iowa, Kentucky, Los Angeles, Louisiana, New Jersey, New Mexico, Seattle-Puget Sound, and Utah. This translates to approximately 26% of African Americans, 41% of Hispanics, 43% of Amer‐ ican Indians and Alaska Natives, 54% of Asians, and 71% of Hawaiian/Pacific Islanders. It is published annually, with 2012 being the most recent year for which data are available.

The following statistical data was obtained from the Surveillance, Epidemiology and End Results (SEER) Program of the National Cancer Institute (NCI), specifically from data previously published in the *SEER Cancer Statistic Review (CSR) 1975-2012*, which was released April 23, 2015. The NCI funds for the program through Centers for Disease Control and Presentation (CDC), National Program of Cancer Registries, and involved states' contributions. The SEER program was conceptualized in 1973, with a mission to report the ―most recent cancer incidence, mortality, survival, prevalence, and lifetime risks statistics. It originally only

**How common is this cancer?** It is estimated that there will be 132,700 new colorectal cancer cases in 2015. [21] This comprises 8% of all new cancer cases (Figure 1). [21] Of those new cancer cases, there will be an estimated 49,700 deaths. [21] This comprises 8.4% of all cancer deaths

**Who gets this cancer?** Colorectal cancer is more common in men than in women. In 2014, there were a total of 135,260 people diagnosed with colorectal cancer: 70,099 men versus 65,161 women. [22] Based on SEER 18, this means that 48.9 per 100,000 persons new cases were male,

Since then, it has expanded to include the following population-based cancer registries: Alaska Native Tumor Registry, Arizona Indians, Cherokee Nation, Connecticut, Detroid, Georgia Center for Cancer Statistics (Atlanta, Greater Georgia, Rural Georgia), Greater Bay Area Cancer Registry (San Francisco-Oakland, San Jose-Monterey), Greater California, Hawaii, Iowa, Kentucky, Los Angeles, Louisiana, New Jersey, New Mexico, Seattle-Puget Sound, and Utah. This translates to approximately 26% of African Americans, 41% of Hispanics, 43% of American Indians and Alaska Natives, 54% of Asians, and 71% of Hawaiian/Pacific Islanders. It is published annually, with 2012 being the most recent year for which data is available. **How common is this cancer?** It is estimated that there will be 132,700 new cases in 2015 (Table 4). This comprises 8% of all new cancer cases. Of those new cases, it is estimated that there will be 49,700 deaths. This comprises 8.4% of all cancer deaths in the United States

tions.

(Table 4).

represented about 10% of the US population.

68 Screening for Colorectal Cancer with Colonoscopy

**Statistical Temporal Trends**

**Statistical Temporal Trends**

represented ~10% of the US population.

Colon and rectum cancer represents **8.0%** of all new cancer cases in the U.S.

**4.2. Temporal trends in the united states**

(Figure 1).

**Figure 1.** Colon and rectum cancer in the U.S.

while 37.1 per 100,000 persons were female. [20]

While colorectal cancer is more common in men than in women, the gender bias is smaller when all races are included. However, the gender bias remains wide when race and ethnicity are factored in. The greatest divide was found in African American males versus females, with 61.2 per 100,000 new cases in black men versus 46.0 per 100,000 new cases in black women. [20]

Other race/ethnicities also showed a divide, but not as wide. Hispanic male new cases were 30/100,000 while female new cases were 43.3/100,000. American Indian/Alaska Native male new cases were 35.7/100,000 while female new cases were 46.3/100,000. Asian/Pacific Islander male new cases were 31.3/100,000 while female new cases were 42.2/100,000. White male new cases were 36.3/100,000 while female new cases were 47.8/100,000 (Table 5). [20]

**At what age is this cancer most frequently diagnosed?** Colorectal cancer is most frequently diagnosed among those aged 65–74 years old. [20] This age group comprises 23.9% of new cases. [20] The median age is 68 years old (Table 6; Figure 2).

**Persons by Race/Ethnicity & Sex**

**Females Males**

**Table 5. Number of New Colon and Rectal Cancer Cases/100,000**

Other race/ethnicities also showed a divide, but not as wide. Hispanic male new cases were 30/100,000, while Hispanic female new cases were 43.3/100,000. American Indican/Alaska Number of new Native males cases were 35.7/100,000, while American Indian/Alaska Native females were 46.3/100,000. Asian/Pacific Islander males were 31.3/100,000 and females were 42.2/100,000. And, number of new white male cases were 36.3/100,000 and white females.

At what age is this cancer most frequently diagnosed? Colorectal cancer is most frequently diagnosed among those aged 65-74, comprising 23.9% of new cases by age group. The Source: SEER 18 2008-2012, Age Adjusted

Source: SEER 18 2008-2012, Age Adjusted

Table 5. Percent of Deaths by Age Group

median age is 68 years old. In the younger age groups (all races, both sexes), percent of new cases were 0.1% (<20 years old), 1.3% (20-34 years old), 14.5% (45-54 years old), and 21.5%

#### Source: SEER 2015

#### **Table 6.** Percent of Deaths by Age Group

Source: SEER 2015

There is different distribution based on age at diagnosis in different gender groups. In women, colon cancer tends to arise in an older population (mean age being 73 years old; Figure 2; in comparison, colon cancer tends to arise in a younger population in men (mean age being 69 years old; Figure 2). [9]

In the younger age groups (all races, both sexes), those <20 years old comprised of 0.1% of new cases; 20–34 years old comprised of 1.3%; 45–54 years old comprised of 14.5%; 55–64 years old comprised of 21.5%.

distribution based on age at diagnosis in gender groups. In women, colon cancer tends to arise in an older population with mean diagnosis being 73 years in men compared to 69 years in men

distribution based on age at diagnosis in gender groups. In women, colon cancer tends to arise in an older population with mean diagnosis being 73 years in men compared to 69 years in men

73 • Female Median Age at Diagnosis The number of new cases will comprise of 42.2 per 100,000 men and women per year, while the number of deaths will comprise of 15.5 per 100,000 men and women per year (age-adjusted **Figure 2.** Median age at which colorectal cancer is most frequently diagnosed.

cases and deaths based on 2008-2012 data).

(Haggard 2014).

being diagnosed with colon or rectal cancer (all races/sexes).

(Haggard 2014).

69

In the older age groups (all races, both sexes), 75–84 years old comprised of 22.6% (75–84 years old) and those >84 years old comprised of 12.1%. [20] **What are the survival rates?** Based on data from SEER 18 2005-2011 (SEER Summary Stage 2000), relative survival statistics show that 64.9% of people survive 5 years or more after being diagnosed with colon or rectal cancer (all races/sexes).

**What are the survival rates?** Based on the data from SEER 18 2005–2011, relative survival statistics show that 64.9% of people survive 5 years or more after being diagnosed with colon or rectal cancer (all races/sexes, Figure 3). [20, 22] The number of new cases will comprise of 42.2 per 100,000 men and women per year, while the number of deaths will comprise of 15.5 per 100,000 men and women per year (age-adjusted cases and deaths based on 2008-2012 data). **64.9%** • **Percent Surviving 5 Years** • **2005-2011**

**What are the survival rates?** Based on data from SEER 18 2005-2011 (SEER Summary Stage 2000), relative survival statistics show that 64.9% of people survive 5 years or more after

**Figure 3.** Relative survival rate of colon or rectal cancer.

At what age is this cancer most frequently diagnosed? Colorectal cancer is most frequently diagnosed among those aged 65-74, comprising 23.9% of new cases by age group. The median age is 68 years old. In the younger age groups (all races, both sexes), percent of new cases were 0.1% (<20 years old), 1.3% (20-34 years old), 14.5% (45-54 years old), and 21.5% (55-64 years old). In the older age groups (all races, both sexes), percent of new cases were

**All Races 37.1**

**African-American <sup>46</sup>**

**White 36.3**

Other race/ethnicities also showed a divide, but not as wide. Hispanic male new cases were 30/100,000, while Hispanic female new cases were 43.3/100,000. American Indican/Alaska Number of new Native males cases were 35.7/100,000, while American Indian/Alaska Native females were 46.3/100,000. Asian/Pacific Islander males were 31.3/100,000 and females were 42.2/100,000. And, number of new white male cases were 36.3/100,000 and white females.

**Table 5. Number of New Colon and Rectal Cancer Cases/100,000**

**Non-Hispanic 38.1**

**Hispanic <sup>30</sup>**

**American Indian/Alaska Native 35.7**

**Asian/Pacific Islander 31.3**

**Females Males**

**49.7**

**43.3**

**42.2**

**46.3**

**47.8**

**48.9**

**61.2**

**Persons by Race/Ethnicity & Sex**

70 Screening for Colorectal Cancer with Colonoscopy

There is different distribution based on age at diagnosis in different gender groups. In women, colon cancer tends to arise in an older population (mean age being 73 years old; Figure 2; in comparison, colon cancer tends to arise in a younger population in men (mean age being 69

In the younger age groups (all races, both sexes), those <20 years old comprised of 0.1% of new cases; 20–34 years old comprised of 1.3%; 45–54 years old comprised of 14.5%; 55–64 years old

22.6% (75-84 years old) and 12.1% (>84 years old). Table 5. Percent of Deaths by Age Group

**Table 5.** Number of New Colon and Rectal Cancer Cases/100,000 Persons by Race/Ethnicity & Sex

Source: SEER 2015

Source: SEER 18 2008-2012, Age Adjusted

Source: SEER 18 2008-2012, Age Adjusted

Source: SEER 2015

**Table 6.** Percent of Deaths by Age Group

years old; Figure 2). [9]

comprised of 21.5%.

**Does staging influence survival rates?** Cancer stage at diagnosis will determine both treat‐ ment options and has a strong influence on the length of survival. Obviously, the earlier the cancer is caught, the better the chance of survival.

Current statistics show that 39.5% of colon and rectal cancers are diagnosed at the local stage (confined to primary site), with a 5-year survival for localized colon and rectal cancer being very high at 90.1% [20] (Table 7).

Thirty-six percent of cancers in the regional stage (those spread to regional lymph nodes) have a 70.8% 5-year relative survival rate. [20] Twenty percent of cancers in the distant stage (those that metastasized) carry a 13.1% 5-year relative survival rate. [20] Lastly, those that are unstaged (5%) have a 34.5% 5-year survival rate [20] (Table 7).

**Table 7.** 5-Year Relative Survival and Percent of Colon and Rectal Cancer Cases by Stage at Diagnosis [20]

**Does the site of cancer change the incidence?** Distribution of colon cancers also vary. This suggests that there are different pathogenic etiologies and carcinogenic mechanisms involved in different sites of the colon (and rectum).

The most common tumor locations in decreasing order are the descending colon (40–42%), rectosigmoid and rectum (30–33%), cecum and ascending colon (25–30%), and transverse colon (10–13%). [22, 23] In other words, 50% of colon cancers are within reach of a flexible sigmoi‐ doscope [24] (Table 8).

**Who dies from this cancer?** As with all cancers, the death rates increase with age. Among both gender groups, it is the second leading cause of cancer deaths — behind lung cancer — with peak incidence being in the seventh decade of life. [2, 20]

In the United States, colorectal cancer is the second leading cause of death. [2] Unfortunately, each year there are >55,000 deaths (26,804 men; 24,979 women). [20]

The percent of deaths is highest among those aged 75–84 at 26.6%. [20] The median age at death is 73 years old (Figure 4). [20] This age group comprises 26.6% of all colorectal cancer deaths [20] (Table 9).

In the younger age groups (all races, both sexes), percent of deaths in those <20 years old comprised of 0% of new cases; 20–34 years old comprised of 0.7%; 35–44 years old comprised of 2.5%; 45–54 years old comprised of 9.3%; 55–64 years old comprised of 17.9%.

In the older age groups (all races, both sexes), percent of deaths in those 65–74 years old comprised of 22.1% and those >84 years old comprised of 21.0%. [20]

Epidemiology of Colorectal Cancer — Incidence, Lifetime Risk Factors Statistics and Temporal Trends http://dx.doi.org/10.5772/61945 73

Who dies from this cancer? As with all cancers, the death rates increase with age. In the United States, as previously stated, colorectal cancer is the second-leading cause of death.

**Table 8.** Incidence Rates of Colon and Rectal Cancer by Location [20] Unfortunately, each year there are >55,000 deaths (26,804 men; 24,979 women). The percent of deaths is highest among those aged 75-84 (26.6%).

Table 8. Percent of Deaths by Age Group

#### (20-34 years old), 2.5% (35-44 years old), 9.3% (45-54 years old), 17.9% (55-64 years old), 22.1% (65-74 years old). In the older age groups (all races/sexes), percent of deaths were 21% **Table 9.** Percent of Deaths by Age Group

**Does the site of cancer change the incidence?** Distribution of colon cancers also vary. This suggests that there are different pathogenic etiologies and carcinogenic mechanisms involved

**Table 7.** 5-Year Relative Survival and Percent of Colon and Rectal Cancer Cases by Stage at Diagnosis [20]

The most common tumor locations in decreasing order are the descending colon (40–42%), rectosigmoid and rectum (30–33%), cecum and ascending colon (25–30%), and transverse colon (10–13%). [22, 23] In other words, 50% of colon cancers are within reach of a flexible sigmoi‐

**Who dies from this cancer?** As with all cancers, the death rates increase with age. Among both gender groups, it is the second leading cause of cancer deaths — behind lung cancer — with

In the United States, colorectal cancer is the second leading cause of death. [2] Unfortunately,

The percent of deaths is highest among those aged 75–84 at 26.6%. [20] The median age at death is 73 years old (Figure 4). [20] This age group comprises 26.6% of all colorectal cancer deaths

In the younger age groups (all races, both sexes), percent of deaths in those <20 years old comprised of 0% of new cases; 20–34 years old comprised of 0.7%; 35–44 years old comprised

In the older age groups (all races, both sexes), percent of deaths in those 65–74 years old

of 2.5%; 45–54 years old comprised of 9.3%; 55–64 years old comprised of 17.9%.

in different sites of the colon (and rectum).

72 Screening for Colorectal Cancer with Colonoscopy

peak incidence being in the seventh decade of life. [2, 20]

each year there are >55,000 deaths (26,804 men; 24,979 women). [20]

comprised of 22.1% and those >84 years old comprised of 21.0%. [20]

doscope [24] (Table 8).

[20] (Table 9).

As more males are diagnosed each year than females, there are more male number of deaths than females. In all races, there were 18.6 number of deaths per 100,000 males versus 13.1 number of deaths per 100,000 females. • Median Age of 73 Death

In the younger age groups (all races/sexes), percent of deaths were 0% (<20 years old), 0.7%

in those >84 years old. The median age at death being 73 (US 2008-2012).

The divide between the genders was even greater when race and ethnicity were factored in. African American males had the highest number of deaths per 100,000: 26.9 (versus 17.8/100,000 females). [25] Males who were identified as non-Hispanic (but not white or black) had the second highest number of deaths (18.9/100,000), followed by American Indian/Alaska native (18.8/100,000) and whites (18.0/100,000). Black females had the higher number of deaths per 100,000 (17.8), followed by American Indian/Alaska native (15.6), non-Hispanic (13.4), and whites (12.7) [20] (Table 10). (18.8/100,000) and whites (18.0/100,000). Black females had the higher number of deaths per 100,000 (17.8), followed by American Indian/Alaska native (15.6), non-Hispanic (13.4), and whites (12.7). **Table 9. Number of Colon and Rectal Cancer Deaths per 100,000**

**Persons by Race/Ethnicity & Sex**

As more males are diagnosed each year than females, there are more male number of deaths

Again African American males had the highest number of deaths per 100,000: 26.9 (versus 17.8/100,000 females). Males who identified as non-Hispanic (but not white or black), had the second highest numbers of deaths (18.9/100,000), followed by American Indian/Alaska native

Source: U.S. 2008-2012, Age-Adjusted Source: U.S. 2008-2012, Age-Adjusted

**Table 10.** Number of Colon and Rectal Cancer Deaths per 100,000 Persons by Race/Ethnicity & Sex

**What are the projection rates of colorectal cancer?** Rates of new colon and rectal cancer diagnosis have been falling each year, over the past 10 years. [26] This is true not only for the United States but also for New Zealand, Australia, and Western Europe.[9] Despite these numbers, the death rate has not changed significantly, however (Table 11).

Source: SEER 9 Incidence & U.S. Mortality 1975-2012, all races/both sexes/rates are age-adjusted

**Table 11.** Incidence & U.S Mortality 1975-2012

### **5. Conclusion**

native (18.8/100,000) and whites (18.0/100,000). Black females had the higher number of deaths per 100,000 (17.8), followed by American Indian/Alaska native (15.6), non-Hispanic (13.4), and

**Table 9. Number of Colon and Rectal Cancer Deaths per 100,000**

**Female Male**

**Non-Hispanic 13.4 18.9 Hispanic 9.6 15.6**

**African-American 17.8 26.9 White 12.7 18 All Races 13.1 18.6**

**American Indian/Alaska Native 15.6 18.8 Asian/Pacific Islander 9.4 13**

**Table 10.** Number of Colon and Rectal Cancer Deaths per 100,000 Persons by Race/Ethnicity & Sex

numbers, the death rate has not changed significantly, however (Table 11).

Source: SEER 9 Incidence & U.S. Mortality 1975-2012, all races/both sexes/rates are age-adjusted

**What are the projection rates of colorectal cancer?** Rates of new colon and rectal cancer diagnosis have been falling each year, over the past 10 years. [26] This is true not only for the United States but also for New Zealand, Australia, and Western Europe.[9] Despite these

**Year 1975 1980 1985 1990 1995 1999 2003 2007**

48.6% 51.1% 58.0% 60.8% 59.7% 64.5% 65.3% 66.5%

As more males are diagnosed each year than females, there are more male number of deaths than females. In all races, there were 18.6 number of deaths per 100,000 males versus 13.1/100,000 females. The divide was even greater when race and ethnicity was factored in. Again African American males had the highest number of deaths per 100,000: 26.9 (versus 17.8/100,000 females). Males who identified as non-Hispanic (but not white or black), had the second highest numbers of deaths (18.9/100,000), followed by American Indian/Alaska native (18.8/100,000) and whites (18.0/100,000). Black females had the higher number of deaths per 100,000 (17.8), followed by American Indian/Alaska native (15.6), non-Hispanic (13.4), and

whites (12.7) [20] (Table 10).

Source: U.S. 2008-2012, Age-Adjusted Source: U.S. 2008-2012, Age-Adjusted

**Table 11.** Incidence & U.S Mortality 1975-2012

**5-Year Relative Survival** whites (12.7).

74 Screening for Colorectal Cancer with Colonoscopy

**Persons by Race/Ethnicity & Sex**

Although new diagnosis rates of colorectal cancer have lowered significantly in both women and men since 1975, more can be done in terms of screening. The drama in these numbers is that colorectal cancer is a preventative cancer, both in screening and in identification of modifiable (i.e., theoretically preventable) risk factors. In fact, if everyone aged 50 years or older had regular screening tests, at least 60% of deaths from this cancer could have been avoided. [3, 19] And with the knowledge that the 5-year survival is close to 90% when colorectal cancer is diagnosed at an early stage, the statistics becomes even more dramatic. **Bottom‐ line:** colorectal cancer is susceptible to screening and aggressive campaigns toward educating the public dictate the future of its incidence and survival.

### **Author details**

Camille Thélin\* and Sanjay Sikka

\*Address all correspondence to: cthelin1@tulane.edu

Department of Internal Medicine, Division of Gastroenterology and Hepatology, Tulane University, New Orleans, LA, United States

#### **References**


www.uspreventiveservicestaskforce.org/Page/SupportingDoc/colorectal-cancerscreening/evaluating-test-strategies-for-colorectal-cancer-screening-a-decision-analy‐ sis-for-the-us-preventive-services-task-force

[5] Nawa T, Kato J, Kawamoto H, Okada H, Yamamoto H, Kohno H, Endo H, Shiratori Y. Differences between right- and left-sided colon cancer in patient characteristics,

[6] American Cancer Society. Colorectal Cancer Facts and Figures 2014-2016. Atlanta: American Cancer Soceity, 2014. Available at http://www.cancer.org/acs/groups/

[8] Siegel RL, Jemal A, Ward EM. Increase in incidence of colorectal cancer among young men and women in the United States. Cancer Epidemiol Biomarkers Prev

[9] Haggar FA, Boushey RP. Colorectal Cancer Epidemiology: Incidence, Mortality, Sur‐ vival, and Risk Factors. Clinics in Colon and Rectal Surgery. 2009;22(4):191–7. doi:

[10] Tárraga L, Pedro J, Albero JS, Rodríguez-Montes JA. "Primary and Secondary Pre‐ vention of Colorectal Cancer." Clinical Medicine Insights. Gastroenterology

[11] Doubeni CA, Laiyema AO, Major JM, et al. Socioeconomic status and the risk of col‐ orectal cancer an analysis of more than a half million adults in the national Insittuaes

[12] Boyle P, Langman JS. ABC of colorectal cancer: Epidemiology. BMJ 2000 Sept

[13] Santarelli R L, Pierre F, Corpet D E. Processed meat and colorectal cancer: a review of epidemiologic and experimental evidence. Nutr Cancer 2008;60(2):131–44

[14] Larsson S C, Wolk A. Meat consumption and risk of colorectal cancer: a meta-analy‐

[15] Pöschl G, Seitz HK. Alcohol and cancer. Alcohol Alcohol 2004 May–Jun;39(3):155–65. [16] Zisman AL, Nickolov A, Brand RE, Gorchow A, Roy HK. Associations between the age at diagnosis and location of colorectal cancer and the use of alcohol and tobacco:

[17] Botteri E, Iodice S, Raimondi S, Maisonneuve P, Lowenfels AB. Cigarette smoking and adenomatous polyps; a meta-analysis. Gastroenterology 2008;134(2);388–96; e3.

[18] Hooker CM, Gallicchio L, Genkinger JM, Comstock GW, Alberg AJ. A prospective cohort study of rectal cancer risk in relation to active cigarette smoking and passive

[19] US Preventative Task Force. Evaluating Test Strategies for Colorectal Cancer Screen‐ ing: A Decision Analysis for the U.S. Preventive Services Task Force: Colorectal Can‐ cer: Screening. November 2014. Available at http://

of Health-AARP Diet and Health Study. Cancer 2012;118:3636–44.

sis of prospective studies. Int J Cancer 2006;119(11):2657–64

implications for screening. Arch Intern Med 2006;166(6):629–34.

smoke exposure. Ann Epidemiol 2008;18:28–35.

cancer morphology and histology. J Gastroenterol Hepatol 2008;23:418–23.

[7] Iacopetta B. Are there two sides to colorectal cancer? Int J Cancer 2002;101:403–8.

content/documents/document/acspc-042280.pdf.

2009;18:1695–8.

76 Screening for Colorectal Cancer with Colonoscopy

10.1055/s-0029/1242458.

30;321(7264):805–8.

2014;7:33–46. PMC. Web. 16 May 2015


## **Basic Endoscopic Findings — Normal and Pathological Findings**

Parth J. Parekh and Sanjay K. Sikka

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61256

#### **Abstract**

Since its inception, colonoscopy has evolved to become the cornerstone for colorectal imaging. The increasing indications for endoscopic evaluation and potential therapeutic intervention parallels technological advances and the expanding diagnostic and thera‐ peutic capabilities of colonoscopy. The diagnostic and therapeutic yield of colonoscopy is highly user dependent. Thus, it is essential for the clinical endoscopist to perform a thor‐ ough endoscopic evaluation and be cognizant of normal and pathologic findings. This re‐ view details normal and pathologic endoscopic findings in a variety of disease states that are often encountered by the clinical endoscopist including colon polyps, inflammatory bowel disease, and infectious and non-infectious colitides. In addition, we review the di‐ agnostic and therapeutic role of colonoscopy in the evaluation of an acute lower gastroin‐ testinal bleed.

**Keywords:** Polyp, pseudopolyp, hyperplastic polyp, adenoma, tubular adenoma, tubulo‐ villous adenoma, sessile adenoma, sessile serrated adenoma, colitis, diverticulosis, hem‐ orrhoids, anal fissure

#### **1. Introduction**

The advent of retrograde colonoscopy in June 1969 revolutionized the field of gastroenterology [1]. It has since evolved to become the gold standard for colorectal imaging [2, 3].

As technology continues to advance, so too does the diagnostic utility and therapeutic capabilities of colonoscopy. Thus, it becomes imperative for the clinical endoscopist to perform a thorough colonoscopic evaluation and be cognizant of normal and pathologic findings as indications for colonoscopy expand. Here, we detail normal and pathologic endoscopic findings in a variety of disease states that are often encountered by the clinical endoscopist

including colon polyps, inflammatory bowel disease (IBD), and infectious and non-infectious colitides. In addition, we review the diagnostic and therapeutic role of colonoscopy in the evaluation of an acute lower gastrointestinal bleed.

#### **2. Polyps and potential progression to colorectal cancer**

Colorectal cancer is the third most common cancer among men and women, and the third leading cause of cancer-related death in the United States [4]. It is estimated that in 2014, 71,830 men and 65,000 women were diagnosed with colorectal cancer with approximately 50,000 mortalities (26,270 men and 24,040 women) as a result of the disease. Globally, colorectal cancer is the fourth leading cause of cancer-related death accounting for approximately 700,000 deaths in 2012 [5]. The vast majority of colorectal cancers stem from benign polyps arising from the mucosal layer. Winawer et al. were among the first to demonstrate that colorectal adenomas have the potential to progress to colorectal adenocarcinoma, thus stressing the importance of colonoscopic polypectomy in colorectal cancer prevention [6]. Subsequent long term data has validated the importance of colonoscopy and colonoscopic polypectomy in the prevention of colorectal cancer-related deaths [7]. To date, colonoscopy remains the cornerstone in colorectal cancer prevention. Unfortunately, the "miss rate" of colonoscopy for colorectal cancer and adenomas larger than 1 cm has been reported to be as high as 6% [8] and 17% [9, 10], respec‐ tively.

Adenomas and hamartomatous polyps, later discussed in depth, are polyps that carry malignant potential. They are indolent in nature, typically growing slowly over the span of a decade or more. There is a direct correlation between the size of the adenoma and its risk of developing future advanced adenomas or carcinoma with studies demonstrating this risk to be as high as 7.7% [11], 15.9% [11], and 19.3% [12], for adenomas <5mm, 5–20mm, and >20mm, respectively.

Chromosomal instability and common point mutations occurring in colorectal cancer-related tumor suppressor genes (e.g., APC, P53) or tumor promoter genes (e.g., K-Ras) architect the progression from benign polyps to colorectal cancer. Figure 1 depicts key point mutations and its impact on morphologic changes of a benign polyp to colorectal cancer. There is, however, considerable genetic and epigenetic heterogeneity resulting in different pathways to tumori‐ genesis [13]. Luo et al. sought to evaluate the effect of these alterations on the progression to colorectal cancer by conducting genome-wide array-based studies and comprehensive data analysis of aberrantly methylated loci in normal colon tissue (n=41), colon adenomas (n=42), and colorectal cancer (n=64) [14]. They identified three classes of cancers and two classes of adenomas, high-frequency methylation and low-frequency methylation based on their DNA methylation patterns. Mutant K-Ras was found in a subset of high-frequency methylated adenomas. In addition, they found the methylation signatures of high-frequency methylation adenomas to be similar to those of cancer with low or intermediate levels of methylation, and low-frequency methylation adenomas to have methylation signatures similar to that of normal colon tissue. These findings demonstrated genome-wide alterations in DNA methylation to occur during the early stages of progression of adenomas to colorectal cancer, and the presence of heterogeneity in tumorigenesis, even at the adenoma step of the process.

**Figure 1.** Key point mutations and its impact on morphologic changes of a benign polyp to colorectal cancer.

#### **3. Polyps and pseudopolyps**

including colon polyps, inflammatory bowel disease (IBD), and infectious and non-infectious colitides. In addition, we review the diagnostic and therapeutic role of colonoscopy in the

Colorectal cancer is the third most common cancer among men and women, and the third leading cause of cancer-related death in the United States [4]. It is estimated that in 2014, 71,830 men and 65,000 women were diagnosed with colorectal cancer with approximately 50,000 mortalities (26,270 men and 24,040 women) as a result of the disease. Globally, colorectal cancer is the fourth leading cause of cancer-related death accounting for approximately 700,000 deaths in 2012 [5]. The vast majority of colorectal cancers stem from benign polyps arising from the mucosal layer. Winawer et al. were among the first to demonstrate that colorectal adenomas have the potential to progress to colorectal adenocarcinoma, thus stressing the importance of colonoscopic polypectomy in colorectal cancer prevention [6]. Subsequent long term data has validated the importance of colonoscopy and colonoscopic polypectomy in the prevention of colorectal cancer-related deaths [7]. To date, colonoscopy remains the cornerstone in colorectal cancer prevention. Unfortunately, the "miss rate" of colonoscopy for colorectal cancer and adenomas larger than 1 cm has been reported to be as high as 6% [8] and 17% [9, 10], respec‐

Adenomas and hamartomatous polyps, later discussed in depth, are polyps that carry malignant potential. They are indolent in nature, typically growing slowly over the span of a decade or more. There is a direct correlation between the size of the adenoma and its risk of developing future advanced adenomas or carcinoma with studies demonstrating this risk to be as high as 7.7% [11], 15.9% [11], and 19.3% [12], for adenomas <5mm, 5–20mm, and >20mm,

Chromosomal instability and common point mutations occurring in colorectal cancer-related tumor suppressor genes (e.g., APC, P53) or tumor promoter genes (e.g., K-Ras) architect the progression from benign polyps to colorectal cancer. Figure 1 depicts key point mutations and its impact on morphologic changes of a benign polyp to colorectal cancer. There is, however, considerable genetic and epigenetic heterogeneity resulting in different pathways to tumori‐ genesis [13]. Luo et al. sought to evaluate the effect of these alterations on the progression to colorectal cancer by conducting genome-wide array-based studies and comprehensive data analysis of aberrantly methylated loci in normal colon tissue (n=41), colon adenomas (n=42), and colorectal cancer (n=64) [14]. They identified three classes of cancers and two classes of adenomas, high-frequency methylation and low-frequency methylation based on their DNA methylation patterns. Mutant K-Ras was found in a subset of high-frequency methylated adenomas. In addition, they found the methylation signatures of high-frequency methylation adenomas to be similar to those of cancer with low or intermediate levels of methylation, and low-frequency methylation adenomas to have methylation signatures similar to that of normal colon tissue. These findings demonstrated genome-wide alterations in DNA methylation to

evaluation of an acute lower gastrointestinal bleed.

80 Screening for Colorectal Cancer with Colonoscopy

tively.

respectively.

**2. Polyps and potential progression to colorectal cancer**

In 2003, the Paris Endoscopic Classification arose to describe polyp morphology [15], which can potentially guide the endoscopist toward its malignancy potential [16–18]. Figure 2 provides a schematic overview of the Paris Endoscopic Classification and Figure 3 provides an endoscopic view of differing polyp morphology under traditional white-light colonoscopy. A recent study by van Doom et al. evaluated the interobserver agreement for the Paris Endoscopic Classification among seven expert endoscopists [19]. The seven expert endoscop‐ ists assessed 85 endoscopic video clips depicting polyps. Afterwards, they underwent a digital training module and then assessed the same 85 polyps again. A calculated Fleiss kappa of 0.42 and a mean pairwise agreement of 67% suggested moderate interobserver agreement among the seven experts. In addition, the proportion of lesions labeled as "flat" lesions ranged between 13–40% (p<0.001). The interobserver agreement did not change significantly after the digital training module, which led the investigators to conclude there to be only moderate interobserver agreement among experts for this classification system and that use of this classification system in daily practice is questionable and unsuitable for comparative endo‐ scopist research. Thus, the need for a simplified classification system is necessary to better aid the clinical endoscopist.

**Figure 2.** The Paris Classification based on polyp appearance.

**Figure 3.** Endoscopic views of differing polyp morphology under traditional white-light colonoscopy: (A) Pedunculat‐ ed polyp, (B) Sessile polyp, (C) Flat polyp.

In addition to traditional white-light colonoscopy, several studies have demonstrated the utility of narrow-band-imaging (NBI) to be useful in adenoma detection [20–23]. Under NBI, adenomas appear to have thicker and higher volumes of microvasculature compared to normal mucosa and hyperplastic polyps, resulting in distinct pit patterns that may increase diagnostic yield [23]. This section will review the morphology and histology, malignant potential, and provide endoscopic and pathologic depictions of different polyp subtypes.

#### **3.1. Adenomas**

interobserver agreement among experts for this classification system and that use of this classification system in daily practice is questionable and unsuitable for comparative endo‐ scopist research. Thus, the need for a simplified classification system is necessary to better aid

**Figure 3.** Endoscopic views of differing polyp morphology under traditional white-light colonoscopy: (A) Pedunculat‐

In addition to traditional white-light colonoscopy, several studies have demonstrated the utility of narrow-band-imaging (NBI) to be useful in adenoma detection [20–23]. Under NBI,

the clinical endoscopist.

82 Screening for Colorectal Cancer with Colonoscopy

**Figure 2.** The Paris Classification based on polyp appearance.

ed polyp, (B) Sessile polyp, (C) Flat polyp.

Adenomatous polyps by definition are dysplastic and thus carry malignant potential. They can further be characterized as being an advanced adenoma, synchronous adenoma, or metachronous adenoma. An advanced adenoma is defined as an adenoma with high-grade dysplasia, an adenoma with a size >10 mm, an adenoma with significant villous components (>25%), or an adenoma with evidence of invasive carcinoma [24]. Synchronous adenomas are polyps that are diagnosed at the same time as an index colorectal cancer and metachranous adenomas are ones diagnosed at least six months before or after the diagnosis of an index colorectal cancer [25]. The diagnosis of synchronous and metachranous adenomas are of utmost importance as it can potentially identify individuals at risk for hereditary conditions, thus impacting therapeutic intervention and screening intervals for relatives [26].

#### *3.1.1. Tubular, villous, and tubulovillous adenomas*

Adenomas are characterized as tubular, villous, or tubulovillous (a mixture of the two) based on their glandular architecture. Tubular adenomas, which account for the vast majority of colon adenomas, are characterized by a network of branching adenomatous epithelium and a tubular component of >75% [16]. Figure 4 depicts a histologic representation of a tubular adenoma in the background of normal colon tissue. Villous adenomas, which account for up to 15% of adenomas, are characterized by long glands that extend straight down to the center of the polyp from its surface with a villous component of >75% [16]. Figure 5 depicts a histologic representation of a villous adenoma in the background of normal colon tissue. Lastly, tubu‐ lovillous adenomas, which account for up to 15% of adenomas, are a mixture of the two previous adenomas with a villous component of anywhere from 26–75%. Figure 6 depicts a histologic representation of a tubulovillous adenoma in the background of normal colon tissue.

The CpG island methylator phenotype (CIMP) pathway is composed of methylated promoter regions of multiple putative tumor suppressor genes occurring in colorectal cancer and also in adenomatous polyps [27]. Kakar et al. examined villous/tubulovillous adenomas (n=32) and tubular adenomas (n=30) for *BRAF*/K-Ras mutations and CIMP-status (characterized by methylation of three or more loci at hMLH1, p16, HIC1, RASSF2, MGMT, MINT1, and MINT31) [28]. They found 44% of villous/tubulovillous to be CIMP-positive compared with 27% of tubular adenomas (p=0.08). In addition, villous/tubulovillous adenomas demonstrated significantly higher methylation rates at MGMT (87% vs. 37%; p<0.01) and RASSF2 (94% vs. 70%; p=0.02) when compared to tubular adenomas. Lastly, CIMP-positive adenomas correlat‐ ed with increased size, right-sided location, and increased villous component in villous/ tubulovillous adenomas. This led the authors to conclude that CIMP status is indicative of size, location, and malignant potential, and that methylation of MGMT and RASSF2 increases as adenomas progress from tubular adenomas to villous/tubulovillous adenomas.

#### *3.1.2. Sessile serrated adenomas, traditional serrated adenomas, and hyperplastic polyps*

Serrated lesions account for approximately 30% of colorectal cancers, arising via the serrated neoplasia pathway characterized by widespread DNA methylation and *BRAF* mutations [29]. They are classified histologically as sessile serrated adenomas/polyps (SSA/Ps), traditional serrated adenomas (TSAs), or hyperplastic polyps, with only SSA/Ps and TSAs carrying malignant potential [30]. SSA/Ps typically lack classic dysplasia, however, those that demon‐ strate foci of classic histologic dysplasia and molecular profiles exhibiting methylation of DNA repair genes (e.g., MLH-1) are thought to be precursor lesions to sporadic unstable microsa‐ tellite (MSI-H) cancers. SSA/Ps also exhibit activation of the *BRAF* oncogene, a feature seen in many sporadic MSI-H cancers [31]. Figure 7 depicts two potential molecular pathways of serrated neoplasia.

SSA/Ps tend to be more prominent in the proximal colon [32] as compared with TSAs [33] and hyperplastic polyps [34], which tend to be more prominent in the rectosigmoid. Thus, expert recommendations are to completely remove all serrated lesions proximal to the sigmoid colon and all serrated lesions in the rectosigmoid >5mm [30]. They may be more difficult to detect

**Figure 5.** Histologic representation of villous adenoma in the background of normal colon tissue.

*3.1.2. Sessile serrated adenomas, traditional serrated adenomas, and hyperplastic polyps*

**Figure 4.** Histologic representation of tubular adenoma in the background of normal colon tissue.

serrated neoplasia.

84 Screening for Colorectal Cancer with Colonoscopy

Serrated lesions account for approximately 30% of colorectal cancers, arising via the serrated neoplasia pathway characterized by widespread DNA methylation and *BRAF* mutations [29]. They are classified histologically as sessile serrated adenomas/polyps (SSA/Ps), traditional serrated adenomas (TSAs), or hyperplastic polyps, with only SSA/Ps and TSAs carrying malignant potential [30]. SSA/Ps typically lack classic dysplasia, however, those that demon‐ strate foci of classic histologic dysplasia and molecular profiles exhibiting methylation of DNA repair genes (e.g., MLH-1) are thought to be precursor lesions to sporadic unstable microsa‐ tellite (MSI-H) cancers. SSA/Ps also exhibit activation of the *BRAF* oncogene, a feature seen in many sporadic MSI-H cancers [31]. Figure 7 depicts two potential molecular pathways of

SSA/Ps tend to be more prominent in the proximal colon [32] as compared with TSAs [33] and hyperplastic polyps [34], which tend to be more prominent in the rectosigmoid. Thus, expert recommendations are to completely remove all serrated lesions proximal to the sigmoid colon and all serrated lesions in the rectosigmoid >5mm [30]. They may be more difficult to detect than conventional adenomatous polyps, in particular SSA/Ps, since they are more likely to be flat lesions, and so recent studies have advocated for a longer withdrawal time to increase serrated lesion detection rates [35, 36].

Serrated lesions have a distinct endoscopic appearance albeit often very subtle. A retrospective analysis of high-resolution endoscopic video clips by Tadepalli et al. analyzed the gross morphologic characteristics of 158 SSPs [37]. They found the most prevalent visual descriptors to be the presence of a mucous cap (which may be yellow or green in white light and red under NBI) (63.9%), rim of debris or bubbles (51.9%), alteration of the contour of a fold (37.3%), and interruption of underlying vascular pattern (32%). Figure 8 depicts an SSP under traditional white-light colonoscopy with a superficial mucous cap, its appearance under NBI, and a histologic representation.

**Figure 6.** Histologic representation of a tubulovillous adenoma in the background of normal colon tissue.

Hyperplastic polyps are the most common non-neoplastic polyps in the colon; however, they are oftentimes grossly indistinguishable from adenomatous polyps. Histologically, hyper‐ plastic polyps resemble normal colonic tissue with the exception of proliferation in the basal portion of the crypt and a characteristic "saw tooth" pattern along the crypt axis [38]. The relationship between diminutive hyperplastic polyps in the left colon and proximal neoplasia has long been a topic of debate with studies producing mixed results [39–42]. Hyperplastic polyps found proximal to the left colon, however, have consistently been shown to carry malignant potential and should be resected [39, 43].

#### **3.2. Hamartomatous polyps**

Hamartomatous polyps are polyps that may grossly resemble normal colonic tissue but are histologically a mixture of tissues growing in disarray. Histologically, they contain mucousfilled glands, retention cysts, abundant connective tissue, and/or chronic eosinophilic infiltra‐

**Figure 7.** Potential molecular pathways of serrated neoplasia.

Hyperplastic polyps are the most common non-neoplastic polyps in the colon; however, they are oftentimes grossly indistinguishable from adenomatous polyps. Histologically, hyper‐ plastic polyps resemble normal colonic tissue with the exception of proliferation in the basal portion of the crypt and a characteristic "saw tooth" pattern along the crypt axis [38]. The relationship between diminutive hyperplastic polyps in the left colon and proximal neoplasia has long been a topic of debate with studies producing mixed results [39–42]. Hyperplastic polyps found proximal to the left colon, however, have consistently been shown to carry

**Figure 6.** Histologic representation of a tubulovillous adenoma in the background of normal colon tissue.

Hamartomatous polyps are polyps that may grossly resemble normal colonic tissue but are histologically a mixture of tissues growing in disarray. Histologically, they contain mucousfilled glands, retention cysts, abundant connective tissue, and/or chronic eosinophilic infiltra‐

malignant potential and should be resected [39, 43].

**3.2. Hamartomatous polyps**

86 Screening for Colorectal Cancer with Colonoscopy

**Figure 8.** A) Sessile serrated polyp with mucosal cap under white-light colonoscopy. (B) Sessile serrated polyp under NBI. (C) Histology of sessile serrated polyp demonstrating expanded crypt proliferative zone, exaggerated architecture in crypt region with basilar crypt dilation, inverted crypts, and a predominance of crypts with minimal cell maturation.

tion [44]. Traditionally, they have been classified as non-neoplastic but several associated polyposis syndromes (e.g., Juvenile Polyposis Coli, Peutz-Jegher Syndrome, Cronkhite Canada Syndrome, and Cowden Syndrome) do carry a predilection towards colorectal cancer and other gastrointestinal malignancies.

Juvenile polyps are a type of hamartomatous polyp characterized by dilated cystic glands rather than an increased number of epithelial cells [44]. They can be found at any age, but as the name implies, are more commonly diagnosed during childhood. They are typically removed due to their propensity to bleed. Peutz-Jegher polyps are a type of hamartomatous polyp characterized by glandular epithelium supported by smooth muscle cells contiguous with the muscularis mucosa. Figure 9 depicts an endoscopic view of a hamartomatous polyp and histologic view of a Peutz-Jegher polyp.

**Figure 9.** Endoscopic view of a hamartomatous polyp and histologic view of a Peutz-Jegher polyp.

#### **3.3. Inflammatory pseudopolyps**

Inflammatory polyps, typically seen in IBD, are indicative of regenerative and/or healing phases of mucosal ulceration and possess no malignant potential. They are formed from discrete islands of residual intact colonic mucosa that result from the ulceration and tissue regeration that is inherent to the disease course [45]. Scattered throughout the colitic region of the colon, they are often numerous, filiform, and can be large enough to encompass the lumen resulting in intussusception or luminal obstruction [45, 46]. The clinical endoscopist ought to be cognizant of clusters of localized giant pseudopolyposis as they may be associated with occult dysplasia [47]. Histologically, inflammatory pseudopolyps are characterized by inflamed lamina propria and distorted colonic epithelium [48]. Surface erosions, congestion, hemorrhage and/or crypt abscesses may also be present [48]. Figure 10 depicts an endoscopic and histologic view of an inflammatory pseudopolyp.

**Figure 10.** Endoscopic and histologic view of an inflammatory pseudopolyp.

#### **4. Colitis**

tion [44]. Traditionally, they have been classified as non-neoplastic but several associated polyposis syndromes (e.g., Juvenile Polyposis Coli, Peutz-Jegher Syndrome, Cronkhite Canada Syndrome, and Cowden Syndrome) do carry a predilection towards colorectal cancer

Juvenile polyps are a type of hamartomatous polyp characterized by dilated cystic glands rather than an increased number of epithelial cells [44]. They can be found at any age, but as the name implies, are more commonly diagnosed during childhood. They are typically removed due to their propensity to bleed. Peutz-Jegher polyps are a type of hamartomatous polyp characterized by glandular epithelium supported by smooth muscle cells contiguous with the muscularis mucosa. Figure 9 depicts an endoscopic view of a hamartomatous polyp

**Figure 9.** Endoscopic view of a hamartomatous polyp and histologic view of a Peutz-Jegher polyp.

Inflammatory polyps, typically seen in IBD, are indicative of regenerative and/or healing phases of mucosal ulceration and possess no malignant potential. They are formed from discrete islands of residual intact colonic mucosa that result from the ulceration and tissue regeration that is inherent to the disease course [45]. Scattered throughout the colitic region of the colon, they are often numerous, filiform, and can be large enough to encompass the lumen resulting in intussusception or luminal obstruction [45, 46]. The clinical endoscopist ought to be cognizant of clusters of localized giant pseudopolyposis as they may be associated with occult dysplasia [47]. Histologically, inflammatory pseudopolyps are characterized by

and other gastrointestinal malignancies.

88 Screening for Colorectal Cancer with Colonoscopy

and histologic view of a Peutz-Jegher polyp.

**3.3. Inflammatory pseudopolyps**

#### **4.1. Inflammatory bowel disease**

In patients with a clinical presentation suggestive of IBD, colonoscopy with ileoscopy can be used to make the initial diagnosis as it allows for direct visualization and biopsy of rectal, colonic, and terminal ileum mucosa [49]. In addition, it can assess disease activity and monitor therapeutic response, provide surveillance of dysplasia or neoplasia, and lastly provide therapeutic intervention such as stricture dilation [49] or closure of fistulae and anastomotic leakages [50].

The use of endoscopic appearance in distinguishing IBD from other non-IBD colitides is limited [51] as there are a number of 'IBD mimickers' including but not limited to colonic tuberculosis [52], Behçet's disease [53], and segmental colitis associated with diverticular disease [54]. In addition to tuberculosis, there are hosts of other infectious colitides that can also endoscopi‐ cally mimic IBD [51, 55]. Table 1 provides an endoscopic description of various infectious colitides. Once these other etiologies have been excluded, colonoscopy can often shed light in distinguishing Crohn's disease (CD) from ulcerative colitis (UC), which is important for disease management. The data gathered from an index colonoscopy is of utmost importance owning to the fact that once therapy is initiated for IBD, discriminating features of CD from UC may be obscured [56, 57].


**Table 1.** Endoscopic description of various infectious colitides [54].

#### *4.1.1. Endoscopic features of UC and Mayo Scoring System*

Endoscopically, classic UC starts in the rectum and progresses proximally, sometimes as far as the ileo-cecal valve, in a circumferential and contiguous fashion with diffused and contin‐ uous inflammation [58]. Endoscopic features suggestive of UC include erythema, edema resulting in a loss of the usual vascular patter, granular appearing mucosa, increased friability, and small superficial erosions and ulcers surrounded by diffuse inflammation [59]. These classic visual features are used to endoscopically score the extent of the disease. The Mayo Scoring System was derived in order provide an objective measure describing the endoscopic extent of the disease. Lemmens et al. sought to evaluate the correlation between endoscsopy and histology with use of the Mayo Scoring System [60]. This retrospective study included 236 biopsy sets from 131 patients with known UC. Endoscopy was performed by IBD specialists and graded using the Mayo Scoring System. Biopsy specimens were analyzed by expert gastrointestinal pathologists using the Geboes and Riley histologic scoring systems. They found that at both extremes, inactive and severely active disease, there was a very high concordance rate. For mild disease, however, there were important differences, as histologic examination seemed to have detected more severe disease than endoscopically suspected, thus stressing the need for a combined histologic and endoscopic scoring system when assessing disease activity. Figure 11 depicts the classic endoscopic appearance of UC in relation to the Mayo Scoring System.

disease management. The data gathered from an index colonoscopy is of utmost importance owning to the fact that once therapy is initiated for IBD, discriminating features of CD from

> *Chlamydia* Perianal abscesses, ulcerations, and fistulae *C. difficile* Pseudomembranes and moderately severe colitis,

*Herpes* Proctitis with ulceration, there may be perianal involvement as well. *Histoplasma* Moderately severely colitis, predominantly right sided

*Mycobacterium* Ileal ulceration, may be transverse or circumferential *Nessieria* Proctitis with ulceration, there may be perianal involvement as well *Salmonella* Friable mucosa, ileal and colonic hemorrhages often

*Schistosoma* Extensive colitis may be segmental, polyps often times

*Shigella* Intense patchy colonic erythema that can also include the

present

present

ileum *Treponema* Proctitis with ulceration, there may be perianal involvement as well

*Yersinia* Patchy colitis with ileal ulceration (apthoid)

Endoscopically, classic UC starts in the rectum and progresses proximally, sometimes as far as the ileo-cecal valve, in a circumferential and contiguous fashion with diffused and contin‐ uous inflammation [58]. Endoscopic features suggestive of UC include erythema, edema resulting in a loss of the usual vascular patter, granular appearing mucosa, increased friability, and small superficial erosions and ulcers surrounded by diffuse inflammation [59]. These classic visual features are used to endoscopically score the extent of the disease. The Mayo

predominantly left sided *Cytomegalovirus* Colitis with ulceration (typically punched out and shallow)

**Infectious Etiology Endoscopic Appearance** *Apergillus* Hemorrhagic ulcerations *Campylobacter* Colonic erythema and ulceration

*Entamoeba* Acute colitis with ulceration *E. coli 0157:H7* Moderately severe colitis

*Klebsiella* Hemorrhagic colitis

**Table 1.** Endoscopic description of various infectious colitides [54].

*4.1.1. Endoscopic features of UC and Mayo Scoring System*

UC may be obscured [56, 57].

90 Screening for Colorectal Cancer with Colonoscopy

**Figure 11.** Classic endoscopic appearance of UC in relation to the Mayo Scoring System.

#### *4.1.2. Endoscopic features of CD and the Simple Endoscopic Score for CD (SES-CD)*

Inflammation in CD can span the entire gastrointestinal tract with nearly 55% of cases involving the terminal ileum and colon, 40% involving exclusively the ileum, and 25% involving the colon alone [61]. Rectal involvement occurs in up to 50% of patients with CD [62]. It should be noted that while terminal ileal involvement is strongly suggestive of CD, it might also occur in patients with UC, particularly pan-colitic UC, by way of "backwash" of cecal contents or "backwash ileitis" [63, 64]. The exact pathogenesis of "backwash ileitis" remains poorly understood, however it is believed that in patients with pan-colitic UC, the terminal ileum becomes inflamed stemming from chronic exposure to cecal contents.

Endoscopically, classic CD appears as "skip lesions" or areas of inflammation interposed between islands of normal mucosa, "cobblestone" appearance of the mucosal surface due to submucosal inflammation and edema, and deep, longitudinal, polycyclic ulcers [55]. In 2004, the SES-CD was derived in order to provide an objective measure describing the endoscopic extent of the disease [65]. To date, prospective data evaluating the utility of SES-CD in predicting corticosteroid-free clinical remission and long-term disease progression is lacking [66, 67]. Figure 12 depicts the classic endoscopic appearance of CD as well as the SES-CD. Table 2 illustrates the key endoscopic differences between UC and CD.


**Figure 12.** Classic endoscopic appearance of CD as well as the SES-CD.


**Table 2.** Key endoscopic differences between UC and CD [54].

#### **4.2. Microscopic (Lymphocytic and collagenous) and eosinophilic colitis**

submucosal inflammation and edema, and deep, longitudinal, polycyclic ulcers [55]. In 2004, the SES-CD was derived in order to provide an objective measure describing the endoscopic extent of the disease [65]. To date, prospective data evaluating the utility of SES-CD in predicting corticosteroid-free clinical remission and long-term disease progression is lacking [66, 67]. Figure 12 depicts the classic endoscopic appearance of CD as well as the SES-CD. Table

2 illustrates the key endoscopic differences between UC and CD.

92 Screening for Colorectal Cancer with Colonoscopy

**Figure 12.** Classic endoscopic appearance of CD as well as the SES-CD.

**Table 2.** Key endoscopic differences between UC and CD [54].

**Endoscopic Features Ulcerative Colitis Crohn's Disease**

**Aphthous Ulcers** √ √√√ **Cobblestone Appearance** x √√ **Deep Ulcers** x √√√ **Erythema** √√√ √√ **Granular Mucosa** √√√ √ **Ileal Ulcers** x √√√ **Loss of Vascular Pattern** √√√ √ **Pseudopolyp** √√√ √√√ **Patchy Inflammation** x √√√ **Rectal Involvement** √√√√ √√

While microscopic colitis by definition is a histologic diagnosis, emerging data suggests that it may not always present with normal endoscopic findings [68–72]. Microscopic colitis is further subdivided into lymphocytic colitis and collagenous colitis depending on the presence of lymphocytic predominant infiltration or collagen deposition, respectively [73]. There have been several macroscopic lesions associated with collagenous colitis including longitudinal ulcers [69,70], hypervascularity [71], loss of normal vascularity [72], and exudative bleeding [73]. A retrospective study by Park et al. sought to investigate macroscopic lesions seen on the endoscopy in 14 patients with diagnosed lymphocytic colitis [68]. Patients with more severe diarrhea demonstrated macroscopic lesions on colonoscopy that included hypervascularity and exudative bleeding, which led to the conclusion that lymphocytic colitis may not always present with a normal endoscopically appearing mucosa. Figure 13 depicts lymphocytic colitis associated with hypervascular mucosa and exudative bleeding.

**Figure 13.** Hypervascular mucosa and exudative bleeding associated with lymphocytic colitis.

Eosinophilic disorders can span the entirety of the gastrointestinal tract, including the esophagus (eosinophilic esophagitis), stomach and small intestine (eosinophilic gastroenteri‐ tis), and the colon (eosinophilic colitis). Eosinophilic colitis is the least frequent manifestation of primary eosinophilic gastrointestinal disorders with only a few reports reported over the last four decades [74]. Secondary eosinophilic colitis can stem from several conditions including parasitic infections (e.g., *Strongyloides stercoralis*[75], *Enterobius vermicularis*[76], and *Trichuris trichiura* [77]), drug-induced (e.g., clozapine [78], carbamazepine [79], rifampicin [80], non-steroidal anti-inflammatory drugs [81, 82], tacrolimius [83], and gold [84]), auto-immune disorders (e.g., scleroderma [85], dermatomyositis and polymyositis [86, 87], and vasculitides (e.g., Churg-Strauss syndrome [88]). Endoscopic features suggestive of eosinophilic colitis include an edematous mucosa with loss of normal vascular pattern, patchy erythema, and superficial ulcerations [74].

#### **4.3. Ischemic colitis**

Ischemic colitis occurs as a result of inadequate blood supply to the large colon, typically affecting the critically ill and elderly population [89]. A recent retrospective study by Church et al. examined the role of urgent bedside colonoscopy in critically ill patients [90]. This study included 41 patients totaling 49 bedside colonoscopies with the most common indication being to exclude ischemic colitis (n=25). Of those 25, the diagnosis was confirmed in 19 with 14 patients subsequently undergoing surgical intervention, which led the authors to conclude that bedside colonoscopy is helpful in the diagnosis of acute lower gastrointestinal disease and can potentially guide therapeutic management in critically ill patients. There are several endoscopic findings that may assist in the diagnosis of ischemic colitis, one of which is the colon single-stripe sign. Zuckerman retrospectively studied 26 patients with endoscopic evidence of the colon single-stripe sign and compared it with 58 consecutive patients without a stripe [91]. All patients in the colon single-strip cohort had a stripe that was >5cm in length predominantly in the left colon (89%). Patients with the colon single-stripe sign were signifi‐ cantly more likely to have evidence of a preceding ischemic event (62%) compared to the colitis comparison group (7%). Histologically, patients with the colon single-stripe sign had micro‐ scopic evidence of ischemic injury compared to the colitis cohort (75% vs. 13%, respectively; p<0.0001). Next, the clinical course and outcome of the 26 patients with the colon single-stripe sign was compared with 22 patients with circumferentially involved ischemic colitis. None of the patients with the colon single-stripe sign required surgical intervention compared with 27% of patients with circumferential ischemic colitis. In addition, mortality rates were higher in the circumferential ischemic colitis group compared with patients with the colon singlestripe sign (41% vs. 4%, respectively; p<0.05). This led the authors to conclude that the colonsingle stripe sign can manifest endoscopically, typically in a milder disease in the clinical spectrum of ischemic colitis [91]. Other endoscopic manifestations of ischemic colitis include petechial hemorrhages, edematous and fragile mucosa, segmental erythema, scattered erosions, and longitudinal ulcerations [92]. The 'watershed areas' areas (e.g., splenic flexure and transverse colon) are areas most vulnerable to ischemia due to the fact that they have the fewest collateral circulation. Figure 14 depicts various endoscopic manifestations of ischemic colitis.

#### **4.4. Graft-Versus-Host Disease (GVHD)**

Acute GVHD is associated with significant morbidity and mortality in the first 100 days following allogeneic hematopoietic progenitor stem cell transplant [93]. Acute GVHD can have GI manifestations (abdominal pain, nausea/vomiting, and diarrhea), obstructive jaundice, or skin rash. Gastroenterologists are often times consulted for endoscopic evaluation to rule out GHVD, when post-transplant patients present with GI manifestations in the absence of liver

**Figure 14.** Various endoscopic manifestations of ischemic colitis.

disorders (e.g., scleroderma [85], dermatomyositis and polymyositis [86, 87], and vasculitides (e.g., Churg-Strauss syndrome [88]). Endoscopic features suggestive of eosinophilic colitis include an edematous mucosa with loss of normal vascular pattern, patchy erythema, and

Ischemic colitis occurs as a result of inadequate blood supply to the large colon, typically affecting the critically ill and elderly population [89]. A recent retrospective study by Church et al. examined the role of urgent bedside colonoscopy in critically ill patients [90]. This study included 41 patients totaling 49 bedside colonoscopies with the most common indication being to exclude ischemic colitis (n=25). Of those 25, the diagnosis was confirmed in 19 with 14 patients subsequently undergoing surgical intervention, which led the authors to conclude that bedside colonoscopy is helpful in the diagnosis of acute lower gastrointestinal disease and can potentially guide therapeutic management in critically ill patients. There are several endoscopic findings that may assist in the diagnosis of ischemic colitis, one of which is the colon single-stripe sign. Zuckerman retrospectively studied 26 patients with endoscopic evidence of the colon single-stripe sign and compared it with 58 consecutive patients without a stripe [91]. All patients in the colon single-strip cohort had a stripe that was >5cm in length predominantly in the left colon (89%). Patients with the colon single-stripe sign were signifi‐ cantly more likely to have evidence of a preceding ischemic event (62%) compared to the colitis comparison group (7%). Histologically, patients with the colon single-stripe sign had micro‐ scopic evidence of ischemic injury compared to the colitis cohort (75% vs. 13%, respectively; p<0.0001). Next, the clinical course and outcome of the 26 patients with the colon single-stripe sign was compared with 22 patients with circumferentially involved ischemic colitis. None of the patients with the colon single-stripe sign required surgical intervention compared with 27% of patients with circumferential ischemic colitis. In addition, mortality rates were higher in the circumferential ischemic colitis group compared with patients with the colon singlestripe sign (41% vs. 4%, respectively; p<0.05). This led the authors to conclude that the colonsingle stripe sign can manifest endoscopically, typically in a milder disease in the clinical spectrum of ischemic colitis [91]. Other endoscopic manifestations of ischemic colitis include petechial hemorrhages, edematous and fragile mucosa, segmental erythema, scattered erosions, and longitudinal ulcerations [92]. The 'watershed areas' areas (e.g., splenic flexure and transverse colon) are areas most vulnerable to ischemia due to the fact that they have the fewest collateral circulation. Figure 14 depicts various endoscopic manifestations of ischemic

Acute GVHD is associated with significant morbidity and mortality in the first 100 days following allogeneic hematopoietic progenitor stem cell transplant [93]. Acute GVHD can have GI manifestations (abdominal pain, nausea/vomiting, and diarrhea), obstructive jaundice, or skin rash. Gastroenterologists are often times consulted for endoscopic evaluation to rule out GHVD, when post-transplant patients present with GI manifestations in the absence of liver

superficial ulcerations [74].

94 Screening for Colorectal Cancer with Colonoscopy

**4.3. Ischemic colitis**

colitis.

**4.4. Graft-Versus-Host Disease (GVHD)**

or dermatologic involvement. In a majority of patients, flexible sigmoidoscopy with rectal biopsies allow for histologic diagnosis of GVHD and thus colonoscopy is not necessary [94, 95]. Endoscopic features of GVHD include diffuse edema, hyperemia, patchy erosions, scattered ulcers, sloughing, and active bleeding [96].

#### **5. Evaluation of Lower Gastrointestinal Bleeding (LGIB)**

The incidence of LGIB is approximately 20 per 100,000, with an associated all cause mortality of 3.9% [97]. The three most common causes of LGIB include angioectasias, diverticular bleeding, and hemorrhoidal bleeding [98]. Colonic ulcerations secondary to underlying IBD or chronic NSAID use, stercoral ulcer, Dieulafoy's lesion, or colorectal varices are less common etiologies of LGIB. In addition, an upper gastrointestinal source should also be included in the differential being that upwards of 15% of patients with severe hematoche‐ zia are found to have an upper gastrointestinal source [99]. In a hemodynamically stable patient, colonoscopy remains the cornerstone in the diagnosis of an LGIB. Figure 15 is a suggested algorithm by Parekh et al. for the role of colonoscopy in the evaluation of a hemodynamically stable LGIB [100].

Diverticulosis of the colon is an out-pouching of colonic mucosa through weakened layers of muscle in the colon wall. The incidence of diverticular increases after the age of 40 [101]. While in itself benign, complications of diverticular disease include diverticulitis, which is the inflammation or infection of diverticula, and painless bleeding, which may be life threatening.

**Figure 15.** Suggested algorithm by Parekh et al. for the role of colonoscopy in the evaluation of a hemodynamically stable LGIB [100].

Therefore, it is important for the endoscopist to inform the patient of symptoms of potential complications of diverticular disease.

Colonic angioectasias, previously referred to as arteriovenous malformations or angiodyspla‐ sias, are a common source of lower gastrointestinal bleeding [102]. They can often times be difficult to identify if not actively bleeding. Figure 16 is an example of colonic diverticula and an angioectasia seen endoscopically.

**Figure 16.** Colonic diverticula and an angioectasia seen endoscopically.

#### **6. Hemorrhoids and anal fissures**

Hemorrhoids are vascular structures in the anal canal that act as cushions to help with stool control [103]. When they become swollen or inflamed, internal hemorrhoids (above the dentate line) can present as painless rectal bleeding. External hemorrhoids can result in pain when thrombosed, or painful bleeding if ulceration occurs from pressure necrosis [103]. Skin tags may be evidence of prior thrombosed external hemorrhoids.

An anal fissure is a linear tear or crack in the distal anal canal. It often presents as painful defecation. Initially it usually involves only the epithelium and progresses to include the full thickness of the anal mucosa. Figure 17 is an example of an internal hemorrhoid, external hemorrhoid, skin tag, and an anal fissure.

**Figure 17.** Internal hemorrhoid, external hemorrhoid, skin tag, and an anal fissure.

#### **7. Conclusion**

Therefore, it is important for the endoscopist to inform the patient of symptoms of potential

**Figure 15.** Suggested algorithm by Parekh et al. for the role of colonoscopy in the evaluation of a hemodynamically

Colonic angioectasias, previously referred to as arteriovenous malformations or angiodyspla‐ sias, are a common source of lower gastrointestinal bleeding [102]. They can often times be difficult to identify if not actively bleeding. Figure 16 is an example of colonic diverticula and

complications of diverticular disease.

96 Screening for Colorectal Cancer with Colonoscopy

stable LGIB [100].

an angioectasia seen endoscopically.

**Figure 16.** Colonic diverticula and an angioectasia seen endoscopically.

Colonoscopy is important in the diagnosis and therapeutic management of several disease states. To date, colonoscopy remains the gold standard in colorectal cancer prevention. It is the cornerstone in the diagnosis and therapeutic management of IBD, particularly with the recent paradigm shift in the therapeutic management of IBD stressing the importance of endoscopic remission in addition to symptomatic remission. In addition, a thorough colono‐ scopic exam can aid in the diagnosis of other non-IBD colitides. In the acute setting, findings during colonoscopy are not only crucial in diagnosing the underlying etiology but also driving therapeutic management. As technology evolves and indications for colonoscopy expand, it becomes increasingly more crucial for the clinical endoscopist to be knowledgeable of normal and pathologic findings during colonoscopy.

#### **Author details**

Parth J. Parekh and Sanjay K. Sikka\*

\*Address all correspondence to: ssikka2@tulane.edu

Department of Internal Medicine, Division of Gastroenterology and Hepatology, Tulane University, New Orleans, LA, USA

#### **References**


[9] Pickhardt PJ, Nugent PA, Mysliwiec PA, Choi JR, Schindler WR. Location of adeno‐ mas missed by optical colonoscopy. Ann Intern Med. 2004;141(5):352-9.

during colonoscopy are not only crucial in diagnosing the underlying etiology but also driving therapeutic management. As technology evolves and indications for colonoscopy expand, it becomes increasingly more crucial for the clinical endoscopist to be knowledgeable of normal

Department of Internal Medicine, Division of Gastroenterology and Hepatology, Tulane

[1] Wolff WI. Colonoscopy: History and development. Am J Gastroenterol. 1989;84(9):

[2] Rex DK, Rahmani EY, Haseman JH, Lemmel GT, Kaster S, Buckley JS. Relative sensi‐ tivity of colonoscopy and barium enema for detection of colorectal cancer in clinical

[3] Winawer SJ, Zauber AG, Fletcher RH, et al. Guidelines for colonoscopy surveillance after polypectomy: A consensus update by the US Multi-Society Task Force on Color‐ ectal Cancer and the American Cancer Society. CA Cancer J Clin. 2006;56(3):143-59.

[4] Siegel R, Desantis C, Jemal A. Colorectal cancer statistics, 2014. CA Cancer J Clin.

[5] Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-tieulent J, Jemal A. Global cancer statis‐

[6] Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer by colono‐ scopic polypectomy. The National Polyp Study Workgroup. N Engl J Med.

[7] Zauber AG, Winawer SJ, O'brien MJ, et al. Colonoscopic polypectomy and long-term

[8] Bressler B, Paszat LF, Chen Z, Rothwell DM, Vinden C, Rabeneck L. Rates of new or missed colorectal cancers after colonoscopy and their risk factors: a population-based

prevention of colorectal-cancer deaths. N Engl J Med. 2012;366(8):687-96.

and pathologic findings during colonoscopy.

\*Address all correspondence to: ssikka2@tulane.edu

practice. Gastroenterology. 1997;112(1):17-23.

tics, 2012. CA Cancer J Clin. 2015;65(2):87-108.

analysis. Gastroenterology. 2007;132(1):96-102.

Parth J. Parekh and Sanjay K. Sikka\*

98 Screening for Colorectal Cancer with Colonoscopy

University, New Orleans, LA, USA

**Author details**

**References**

1017-25.

2014;64(2):104-17.

1993;329(27):1977-81.


[36] Anderson JC, Butterly LF, Goodrich M, Robinson CM, Weiss JE. Differences in detec‐ tion rates of adenomas and serrated polyps in screening versus surveillance colonos‐ copies, based on the new hampshire colonoscopy registry. Clin Gastroenterol Hepatol. 2013;11(10):1308-12.

[22] East JE, Ignjatovic A, Suzuki N, et al. A randomized, controlled trial of narrow-band imaging vs. high-definition white light for adenoma detection in patients at high risk

[23] Mizuno K, Kudo SE, Ohtsuka K, et al. Narrow-banding images and structures of mi‐

[24] Winawer SJ, Zauber AG. The advanced adenoma as the primary target of screening.

[25] Mattar M, Frankel P, David D, et al. Clinicopathologic significance of synchronous and metachronous adenomas in colorectal cancer. Clin Colorectal Cancer. 2005;5(4):

[26] Karlitz JJ, Hsieh MC, Liu Y, et al. Population-based lynch syndrome screening by mi‐ crosatellite instability in patients ≤50: Prevalence, testing determinants, and result

[27] Worthley DL, Leggett BA. Colorectal cancer: Molecular features and clinical opportu‐

[28] Kakar S, Deng G, Cun L, Sahai V, Kim YS. CpG island methylation is frequently present in tubulovillous and villous adenomas and correlates with size, site, and vil‐

[29] Rosty C, Hewett DG, Brown IS, Leggett BA, Whitehall VL. Serrated polyps of the large intestine: current understanding of diagnosis, pathogenesis, and clinical man‐

[30] Rex DK, Ahnen DJ, Baron JA, et al. Serrated lesions of the colorectum: Review and recommendations from an expert panel. Am J Gastroenterol. 2012;107(9):1315-29.

[31] Spring KJ, Zhao ZZ, Karamatic R, et al. High prevalence of sessile serrated adenomas with BRAF mutations: A prospective study of patients undergoing colonoscopy. Gas‐

[32] Sweetser S, Smyrk TC, Sinicrope FA. Serrated colon polyps as precursors to colorec‐

[33] Anderson JC. Pathogenesis and management of serrated polyps: Current status and

[34] Weston AP, Campbell DR. Diminutive colonic polyps: Histopathology, spatial distri‐ bution, concomitant significant lesions, and treatment complications. Am J Gastroen‐

[35] Butterly L, Robinson CM, Anderson JC, et al. Serrated and adenomatous polyp detec‐ tion increases with longer withdrawal time: Results from the New Hampshire Colo‐

of adenomas. Colorectal Dis. 2012;14(11):e771-8.

Gastrointest Endosc Clin N Am. 2002;12(1):1-9, v.

nities. Clin Biochem Rev. 2010;31(2):31-8.

lous component. Hum Pathol. 2008;39(1):30-6.

agement. J Gastroenterol. 2013;48(3):287-302.

tal cancer. Clin Gastroenterol Hepatol. 2013;11(7):760-7.

noscopy Registry. Am J Gastroenterol. 2014;109(3):417-26.

future directions. Gut Liver. 2014;8(6):582-9.

troenterology. 2006;131(5):1400-7.

terol. 1995;90(1):24-8.

274-8.

100 Screening for Colorectal Cancer with Colonoscopy

crovessels of colonic lesions. Dig Dis Sci. 2011;56(6):1811-7.

availability prior to colon surgery. Am J Gastroenterol. 2015.


[65] Daperno M, D'haens G, Van assche G, et al. Development and validation of a new, simplified endoscopic activity score for Crohn's disease: The SES-CD. Gastrointest Endosc. 2004;60(4):505-12.

[50] Sulz MC, Bertolini R, Frei R, Semadeni GM, Borovicka J, Meyenberger C. Multipur‐ pose use of the over-the-scope-clip system ("Bear claw") in the gastrointestinal tract: Swiss experience in a tertiary center. World J Gastroenterol. 2014;20(43):16287-92.

[51] Fefferman DS, Farrell RJ. Endoscopy in inflammatory bowel disease: Indications, sur‐ veillance, and use in clinical practice. Clin Gastroenterol Hepatol. 2005;3(1):11-24.

[52] Chatzicostas C, Koutroubakis IE, Tzardi M, Roussomoustakaki M, Prassopoulos P, Kouroumalis EA. Colonic tuberculosis mimicking Crohn's disease: Case report. BMC

[53] Lee SK, Kim BK, Kim TI, Kim WH. Differential diagnosis of intestinal Behçet's dis‐ ease and Crohn's disease by colonoscopic findings. Endoscopy. 2009;41(1):9-16.

[54] Del val JH. Old-age inflammatory bowel disease onset: A different problem? World J

[55] Rameshshanker R, Arebi N. Endoscopy in inflammatory bowel disease when and

[56] Kim B, Barnett JL, Kleer CG, Appelman HD. Endoscopic and histological patchiness

[57] Bernstein CN, Shanahan F, Anton PA, Weinstein WM. Patchiness of mucosal inflam‐ mation in treated ulcerative colitis: A prospective study. Gastrointest Endosc.

[58] Jevon GP, Madhur R. Endoscopic and histologic findings in pediatric inflammatory

[59] Tontini GE, Vecchi M, Pastorelli L, Neurath MF, Neumann H. Differential diagnosis in inflammatory bowel disease colitis: State of the art and future perspectives. World

[60] Lemmens B, Arijs I, Van assche G, et al. Correlation between the endoscopic and his‐ tologic score in assessing the activity of ulcerative colitis. Inflamm Bowel Dis.

[61] Freeman HJ. Natural history and clinical behavior of Crohn's disease extending be‐

[62] Nikolaus S, Schreiber S. Diagnostics of inflammatory bowel disease. Gastroenterolo‐

[63] Kaiser AM. Discussion of "Backwash ileitis is strongly associated with colorectal car‐

[64] Haskell H, Andrews CW, Reddy SI, et al. Pathologic features and clinical significance of "backwash" ileitis in ulcerative colitis. Am J Surg Pathol. 2005;29(11):1472-81.

in treated ulcerative colitis. Am J Gastroenterol. 1999;94(11):3258-62.

bowel disease. Gastroenterol Hepatol (N Y). 2010;6(3):174-80.

yond two decades. J Clin Gastroenterol. 2003;37(3):216-9.

cinoma in ulcerative colitis". Gastroenterology. 2002;122(1):245-6.

Gastroenterol. 2002;2(1):10.

102 Screening for Colorectal Cancer with Colonoscopy

1995;42(3):232-7.

2013;19(6):1194-201.

gy. 2007;133(5):1670-89.

Gastroenterol. 2011;17(22):2734-9.

J Gastroenterol. 2015;21(1):21-46.

why. World J Gastrointest Endosc. 2012;4(6):201-11.


effective in diagnosis as upper endoscopy combined with lower endoscopy? Pediatr Blood Cancer. 2013;60(11):1798-800.

[96] Xu CF, Zhu LX, Xu XM, Chen WC, Wu DP. Endoscopic diagnosis of gastrointestinal graft-versus-host disease. World J Gastroenterol. 2008;14(14):2262-7.

[80] Lange P, Oun H, Fuller S, Turney JH. Eosinophilic colitis due to rifampicin. Lancet.

[81] Bridges AJ, Marshall JB, Diaz-arias AA. Acute eosinophilic colitis and hypersensitivi‐ ty reaction associated with naproxen therapy. Am J Med. 1990;89(4):526-7.

[82] Jiménez-sáenz M, González-cámpora R, Linares-santiago E, Herrerías-gutiérrez JM. Bleeding colonic ulcer and eosinophilic colitis: a rare complication of nonsteroidal

[83] Saeed SA, Integlia MJ, Pleskow RG, et al. Tacrolimus-associated eosinophilic gastro‐ enterocolitis in pediatric liver transplant recipients: role of potential food allergies in

[84] Martin DM, Goldman JA, Gilliam J, Nasrallah SM. Gold-induced eosinophilic entero‐ colitis: Response to oral cromolyn sodium. Gastroenterology. 1981;80(6):1567-70. [85] Clouse RE, Alpers DH, Hockenbery DM, Deschryver-kecskemeti K. Pericrypt eosino‐ philic enterocolitis and chronic diarrhea. Gastroenterology. 1992;103(1):168-76.

[86] Barbie DA, Mangi AA, Lauwers GY. Eosinophilic gastroenteritis associated with sys‐

[87] Ahmad M, Soetikno RM, Ahmed A. The differential diagnosis of eosinophilic esoph‐

[88] Avgerinos A, Bourikas L, Tzardi M, Koutroubakis IE. Eosinophilic gastroenteritis as‐ sociated with Churg-Strauss syndrome. Ann Gastroenterol. 2012;25(2):164.

[89] Huguier M, Barrier A, Boelle PY, Houry S, Lacaine F. Ischemic colitis. Am J Surg.

[90] Church J, Kao J. Bedside colonoscopy in intensive care units: indications, techniques,

[91] Zuckerman GR, Prakash C, Merriman RB, Sawhney MS, Deschryver-kecskemeti K, Clouse RE. The colon single-stripe sign and its relationship to ischemic colitis. Am J

[92] Zou X, Cao J, Yao Y, Liu W, Chen L. Endoscopic findings and clinicopathologic char‐ acteristics of ischemic colitis: A report of 85 cases. Dig Dis Sci. 2009;54(9):2009-15. [93] Sultan M, Ramprasad J, Jensen MK, Margolis D, Werlin S. Endoscopic diagnosis of pediatric acute gastrointestinal graft-versus-host disease. J Pediatr Gastroenterol

[94] Aslanian H, Chander B, Robert M, et al. Prospective evaluation of acute graft-versus-

[95] Crowell KR, Patel RA, Fluchel M, Lowichik A, Bryson S, Pohl JF. Endoscopy in the diagnosis of intestinal graft-versus-host disease: Is lower endoscopy with biopsy as

temic lupus erythematosus. J Clin Gastroenterol. 2004;38(10):883-6.

anti-inflammatory drugs. J Clin Gastroenterol. 2006;40(1):84-5.

pathogenesis. Pediatr Transplant. 2006;10(6):730-5.

agitis. J Clin Gastroenterol. 2000;30(3):242-4.

and outcomes. Surg Endosc. 2014;28(9):2679-82.

Gastroenterol. 2003;98(9):2018-22.

host disease. Dig Dis Sci. 2012;57(3):720-5.

Nutr. 2012;55(4):417-20.

2006;192(5):679-84.

1994;344(8932):1296-7.

104 Screening for Colorectal Cancer with Colonoscopy


**Section 3**

## **Looking Ahead**

## **Building up the Future of Colonoscopy – A Synergy between Clinicians and Computer Scientists**

Jorge Bernal, F. Javier Sánchez, Cristina Rodríguez de Miguel and Gloria Fernández-Esparrach

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/61012

#### **Abstract**

Recent advances in endoscopic technology have generated an increasing interest in strengthening the collaboration between clinicians and computers scientist to develop intelligent systems that can provide additional information to clinicians in the differ‐ ent stages of an intervention. The objective of this chapter is to identify clinical draw‐ backs of colonoscopy in order to define potential areas of collaboration. Once areas are defined, we present the challenges that colonoscopy images present in order com‐ putational methods to provide with meaningful output, including those related to im‐ age formation and acquisition, as they are proven to have an impact in the performance of an intelligent system. Finally, we also propose how to define valida‐ tion frameworks in order to assess the performance of a given method, making an special emphasis on how databases should be created and annotated and which met‐ rics should be used to evaluate systems correctly.

**Keywords:** Intelligent systems, Image properties, Validation, Clinical drawbacks, En‐ doluminal scene description

#### **1. Introduction**

#### **1.1. Motivation**

During the last few years there has been an increasing effort in exploring the use of intelligent systems to assist and provide additional information to clinicians in the different stages of an intervention. In this context, we can find in the literature systems aiming at assisting the clinician in in-vivo diagnosis such as KARDIO proposed in [1], which can automatically analyze electrocardiograms, or methods that provide with data to help in the detection and

© 2015 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

diagnosis of breast [2] or prostate cancer [3]. The spread use of Computed Tomography has elicited a new set of methods that help clinicians in intervention planning as exposed in [4]. For instance, we can find systems which allow clinicians to follow the fastest and safest way to target a pulmonary lesion [5], perform laparoscopic surgery [6] or systems such as [7] in the domain of transcatheter aortic valve implantations. However, there is scarce experience with intelligent systems applied to endoscopy where there are only a few methods such as the works presented in [8] in the context of colonoscopy quality assessment which analyzes how clinical procedures have been performed to provide quality scores.

Endoscopic technology has rapidly evolved in the last decade and current equipment allows clinicians to observe the whole endoluminal scene in high definition and, moreover, makes it possible to get different views of the same scene for further analysis by applying automatic techniques of chromoendoscopy [9] as narrow band imaging (NBI) –proposed in [10]-, the Fujinon Intelligent Chromo-Endoscopy (FICE) presented in [11] or Pentax I-scan, which was published in [12]. These advances in endoscopy imaging have generated an increasing interest in strengthening partnerships between clinicians and computer scientists to build applications that can solve some of the challenges that colonoscopy procedures still present nowadays.

It is clear that this potential collaboration between these two domains of knowledge needs from each part to acknowledge the challenges that the analysis of colonoscopy images present related to their area of expertise. Related to this, clinicians need to identify which of the existing drawbacks could be mitigated with the aid of image processing tools and computer scientists must define clearly what can be achieved by means of image processing to provide clinicians with feasible and clinically applicable solutions. Endoscopy imaging analysis present some challenges that are not limited to the ones that the characterization of anatomical structures for detection or diagnosis purposes present; aspects that are rarely covered by existing methods such as image acquisition and formation should be considered as they are proven to have an impact on the output of a given method [13].

Considering this, the focus of this chapter is to present new advances on computer vision methods for colonoscopy and to identify potential clinical issues that may be solved with the aid of computer vision. As it can be observed, this chapter is not written from either a pure clinical or technical point of view but as a way to couple the necessities and challenges of each of the domains in order to build up feasible and clinically applicable systems.

### **2. Introduction to colonoscopy challenges**

#### **2.1. A brief history of endoscopy**

The history of endoscopy, as stated in [14], starts in 1805 with P. Bozzini and his attempts to construct a cystoscope (See Table 1). Although this first endoscope was considered as having failed, the principles incorporated in its design - a light source, a reflective surface (lens) and a series of specula (mirrors)- are the basis of current endoscopes. The technical challenges posed since then have been overcome with the collaboration of physicians, engineers, scientists and optical experts among others. The progress has been slow but constant and initially rigid instruments have been changed by flexible endoscopes; candles and lamps have been replaced by electric filaments and, for vision, single lenses have been supplanted by optic fibers.

diagnosis of breast [2] or prostate cancer [3]. The spread use of Computed Tomography has elicited a new set of methods that help clinicians in intervention planning as exposed in [4]. For instance, we can find systems which allow clinicians to follow the fastest and safest way to target a pulmonary lesion [5], perform laparoscopic surgery [6] or systems such as [7] in the domain of transcatheter aortic valve implantations. However, there is scarce experience with intelligent systems applied to endoscopy where there are only a few methods such as the works presented in [8] in the context of colonoscopy quality assessment which analyzes how clinical

Endoscopic technology has rapidly evolved in the last decade and current equipment allows clinicians to observe the whole endoluminal scene in high definition and, moreover, makes it possible to get different views of the same scene for further analysis by applying automatic techniques of chromoendoscopy [9] as narrow band imaging (NBI) –proposed in [10]-, the Fujinon Intelligent Chromo-Endoscopy (FICE) presented in [11] or Pentax I-scan, which was published in [12]. These advances in endoscopy imaging have generated an increasing interest in strengthening partnerships between clinicians and computer scientists to build applications that can solve some of the challenges that colonoscopy procedures still present nowadays.

It is clear that this potential collaboration between these two domains of knowledge needs from each part to acknowledge the challenges that the analysis of colonoscopy images present related to their area of expertise. Related to this, clinicians need to identify which of the existing drawbacks could be mitigated with the aid of image processing tools and computer scientists must define clearly what can be achieved by means of image processing to provide clinicians with feasible and clinically applicable solutions. Endoscopy imaging analysis present some challenges that are not limited to the ones that the characterization of anatomical structures for detection or diagnosis purposes present; aspects that are rarely covered by existing methods such as image acquisition and formation should be considered as they are proven to

Considering this, the focus of this chapter is to present new advances on computer vision methods for colonoscopy and to identify potential clinical issues that may be solved with the aid of computer vision. As it can be observed, this chapter is not written from either a pure clinical or technical point of view but as a way to couple the necessities and challenges of each

The history of endoscopy, as stated in [14], starts in 1805 with P. Bozzini and his attempts to construct a cystoscope (See Table 1). Although this first endoscope was considered as having failed, the principles incorporated in its design - a light source, a reflective surface (lens) and a series of specula (mirrors)- are the basis of current endoscopes. The technical challenges posed since then have been overcome with the collaboration of physicians, engineers, scientists

of the domains in order to build up feasible and clinically applicable systems.

procedures have been performed to provide quality scores.

110 Screening for Colorectal Cancer with Colonoscopy

have an impact on the output of a given method [13].

**2. Introduction to colonoscopy challenges**

**2.1. A brief history of endoscopy**



**Table 1.** Evolution of endoscopy as a result of collaboration of different disciplines

Shortly after having successfully traversed the esophagus and reached the stomach, the assessment of the duodenum, small intestine and colon were the next steps that were pro‐ gressively addressed and achieved. Other needs were also identified and solved: first, the evolution from diagnostic to operating endoscopes that allowed obtaining biopsies; second, the need of preserving the image of the lesion which was observed. The latter not only reflected clinical needs but also documentation and educational requirements. At that point, several corporations became involved in the development of endoscopic instrumentation and they also designed cameras specifically for endoscopic usage.

Once the fiber optic endoscope was established as a reality by late 1960s, numerous design modifications were performed with the collaboration of physicians in order to augment the utility of the device and increase its resolution. The decade of 1970 witnessed a series of rapid technological advances where a number of instrumental manufactures including ACMI, Olympus Optical Company and Machida Endoscope Company included a variety of innova‐ tions (length, flexibility, channel size...) that improved the performance of the instrument. In 1983 video endoscopy was introduced as the logical consequence of technical advances in microelectronics and all current endoscopes are based on this technology. Video endoscopy allows an easy exploration, instant image acquisition and further storage confirming its utility not only for clinical practice but also for educational purposes.

#### **2.2. High definition endoscopy (The quality of image matters)**

**Year Authorship Development**

Development of the first semiflexible gastroscope. Schindler is considered the

Development of the operating gastroscope by incorporating both a biopsy

forceps and a suction tube within the gastroscope itself.

(Engineer) Design of a miniature gastrocamera that can be introduced into the stomach.

Introduction of the first fiber optic gastroscope.

founder of modern endoscopy.

**<sup>1940</sup>** Cameron Surgical Co. The first flexible gastroscope is made in the USA: the *Cameron Schindler*

James Watson (Physician) Production of a viable endoscopic photographic system.

Olympus Optical Co. Development of the first prototypes of flexible colonoscopes.

Hiromi Shinya (Surgeon) Performance of the first polypectomy with a wire loop snare.

**<sup>1983</sup>** Welch Allyn Inc. Development of an electronic sensor or charge coupled device that is inserted at the tip of the endoscope.

Shortly after having successfully traversed the esophagus and reached the stomach, the assessment of the duodenum, small intestine and colon were the next steps that were pro‐ gressively addressed and achieved. Other needs were also identified and solved: first, the evolution from diagnostic to operating endoscopes that allowed obtaining biopsies; second, the need of preserving the image of the lesion which was observed. The latter not only reflected clinical needs but also documentation and educational requirements. At that point, several corporations became involved in the development of endoscopic instrumentation and they

Once the fiber optic endoscope was established as a reality by late 1960s, numerous design modifications were performed with the collaboration of physicians in order to augment the utility of the device and increase its resolution. The decade of 1970 witnessed a series of rapid technological advances where a number of instrumental manufactures including ACMI, Olympus Optical Company and Machida Endoscope Company included a variety of innova‐

(Gastroenterologist) Description of the first magnifying colonoscope.

**Table 1.** Evolution of endoscopy as a result of collaboration of different disciplines

also designed cameras specifically for endoscopic usage.

*Endoscope*.

Rudolf Schindler (Gastroenterologist) Georg Wolf (Manufacturer)

112 Screening for Colorectal Cancer with Colonoscopy

Edward B. Benedict

Basil Hirschowitz (Gastroenterologist) Larry Curtiss (Physicist)

**<sup>1960</sup>** Machida Endoscope Co.

**<sup>1971</sup>** William I. Wolff (Surgeon)

Masahiro Tada

**2002** Olympus Co. HD endoscopes

(Surgeon)

**<sup>1948</sup>** Harry Segal (Physician)

**<sup>1952</sup>** Tatsuno Uji

**1932**

**1948**

**1957**

**1975**

In the last years, most of the developments in endoscopy have been focused on improving the quality of images, as it is the case of high definition (HD) endoscopes that use a 1080-line television and a high resolution charge coupled device with up to 1.3 million pixels. This allows the acquisition and storage of images with double the resolution of normal television. Other capabilities available in some endoscopes are the following:


**Figure 1.** Example of a same polyp observed with white light (a) and NBI (b).

HD endoscopes (particularly those with magnification function) facilitate the demonstration of the mucosal architectural and vascular patterns that are altered in dysplastic lesions as it can be observed in Fig.2. With regards to the detection rate of lesions, although it is logical to assume that a higher resolution endoscope could provide better results, the results of several studies [15, 16] do not support this hypothesis.

**Figure 2.** Example of a colonoscopy frame observed with conventional endoscope (a) and with high definition endo‐ scope (b).

#### **2.3. The problem of colonic polyps**

Colorectal cancer (CRC) is a serious health problem in the general population and it is consid‐ ered that at least two thirds of CRC develop through the adenoma–carcinoma pathway. Consequently, screening with colonoscopy for CRC and its precursor lesion has become an increasingly practice, as shown in [17]. Several actions have been proposed to optimize colonoscopy such as ensuring colon perfect preparation and carrying out a thorough examina‐ tionofthemucosa which wouldimply a longer withdrawalinspectiontime, as indicatedin[18].

However, colonoscopy still presents some drawbacks being the most relevant the polyp missrate -reported to be as high as 22%- resulting in a lack of total effectiveness [19]. The rate of polyps missed increases significantly in smaller sized polyps (2% for adenomas ≥ 10 mm versus 26% for adenomas < 5 mm) and this has a clinical impact, not only because the prevalence of high-grade dysplasia increases with the size as exposed in [20] but because of the risk of having an interval cancer. Interval colorectal cancers are described as cancers occurring after a negative screening test or examination and they are an important indicator of the quality and effectiveness of CRC screening and surveillance, as stated in [21].

The diagnosis of dysplasia has practical consequences on the management of polyps. There is general consensus on removing all polyps detected during colonoscopy but size is a limiting factor for endoscopic polypectomy. Therefore, having a histological diagnostic of presumption is very useful in order to make the decision of performing or not a polypectomy. In this regard, there are several classifications (NICE, Kudo...) that predict the histology of the lesion based on the characteristics of the image. Kudo [22] proposes a gross classification of pit patterns into 7 types: type I and II pit patterns are characteristic of non-neoplastic lesions such as normal mucosa or hyperplastic polyps whereas pattern types IIIS, IIIL, IV, and a subset of VI are intramucosal neoplastic lesions such as adenoma or intramucosal carcinoma and lesions with a type VN pattern and a subset of type VI suggest deep invasive carcinoma (see Fig. 3).

Building up the Future of Colonoscopy – A Synergy between Clinicians and Computer Scientists http://dx.doi.org/10.5772/61012 115

(a) (b)

**Figure 2.** Example of a colonoscopy frame observed with conventional endoscope (a) and with high definition endo‐

Colorectal cancer (CRC) is a serious health problem in the general population and it is consid‐ ered that at least two thirds of CRC develop through the adenoma–carcinoma pathway. Consequently, screening with colonoscopy for CRC and its precursor lesion has become an increasingly practice, as shown in [17]. Several actions have been proposed to optimize colonoscopy such as ensuring colon perfect preparation and carrying out a thorough examina‐ tionofthemucosa which wouldimply a longer withdrawalinspectiontime, as indicatedin[18].

However, colonoscopy still presents some drawbacks being the most relevant the polyp missrate -reported to be as high as 22%- resulting in a lack of total effectiveness [19]. The rate of polyps missed increases significantly in smaller sized polyps (2% for adenomas ≥ 10 mm versus 26% for adenomas < 5 mm) and this has a clinical impact, not only because the prevalence of high-grade dysplasia increases with the size as exposed in [20] but because of the risk of having an interval cancer. Interval colorectal cancers are described as cancers occurring after a negative screening test or examination and they are an important indicator of the quality and

The diagnosis of dysplasia has practical consequences on the management of polyps. There is general consensus on removing all polyps detected during colonoscopy but size is a limiting factor for endoscopic polypectomy. Therefore, having a histological diagnostic of presumption is very useful in order to make the decision of performing or not a polypectomy. In this regard, there are several classifications (NICE, Kudo...) that predict the histology of the lesion based on the characteristics of the image. Kudo [22] proposes a gross classification of pit patterns into 7 types: type I and II pit patterns are characteristic of non-neoplastic lesions such as normal mucosa or hyperplastic polyps whereas pattern types IIIS, IIIL, IV, and a subset of VI are intramucosal neoplastic lesions such as adenoma or intramucosal carcinoma and lesions with a type VN pattern and a subset of type VI suggest deep invasive carcinoma (see Fig. 3).

effectiveness of CRC screening and surveillance, as stated in [21].

scope (b).

**2.3. The problem of colonic polyps**

114 Screening for Colorectal Cancer with Colonoscopy

**Figure 3.** Examples of Kudo neoplastic lesion classification: (a) Type I; (b) Type II; (c) Type IIIL; (d) Type IIIS; (e) Type IV and (f) Type V.

As this classification applies for magnification endoscopy, when it is used with conventional endoscopy the results are worse. Contrarily, NICE is an international classification of colorectal tumors on the basis of NBI observation either with or without use of a magnifying endoscope [23]. NICE is a simple categorical classification defining three different types based on three characteristics: (i) lesion color; (ii) micro vascular architecture; and (iii) surface pattern. Type 1 is considered an index for hyperplastic lesions, type 2 an index for adenoma or mucosal/ submucosal scanty invasive carcinoma, and type 3 an index for deeply submucosal-invasive carcinoma The problem with these classifications is that diagnostic derives from a subjective visual analysis and requires specific training and a high degree of experience.

Finally, the precise location of the polyps is another meaningful drawback of colonoscopy, not only when planning a surgery but also during successive colonoscopies. This limitation is especially remarkable in the presence of several polyps. In this case, an exhaustive analysis of the surface and boundaries of the polyp could be very helpful.

#### **2.4. Identification of potential collaborative research areas between clinicians and computer scientists**

Considering the mentioned drawbacks of colonoscopy, three potential areas in which com‐ puter science may play a role have been identified:


## **3. Image processing challenges for the analysis of colonoscopy videos**

In order to provide clinicians with meaningful applications, the content of colonoscopy videos and frames must be thoroughly analyzed by computer scientists to search for lesions or indicators defined by clinicians. In this context, the majority of the literature has been focused on developing methods to characterize accurately the different elements of the endoluminal scene, paying special attention to polyps. Although it is clear that anatomical landmarks recognition is essential for application development, the acquisition and generation of high quality images is also crucial for computer vision methods in order to work as they are intended. For instance, the presence of image artifacts has been proven to have an impact in the performance of polyp localization methods, as shown in [13].

Considering this we present in this section a summary of the most important challenges that a given computer vision method must face in order to provide with efficient support to clinicians. We have divided the challenges in two groups: those related to image acquisition and formation and those related to the characterization of anatomical structures needed to build up the clinicians' support system. **3.1 IDENTIFICATION OF ENDOSCOPY IMAGE PARTICULARITIES WITH IMPACT IN IMAGE PROCESSING**

#### **3.1. Identification of endoscopy image particularities with impact in image processing analysis ANALYSIS**

build up the clinicians' support system.

**2.4. Identification of potential collaborative research areas between clinicians and computer**

Considering the mentioned drawbacks of colonoscopy, three potential areas in which com‐

**•** Automatic polyp detection and localization: one of the exposed drawbacks is related to the difficulty on detecting certain types of polyps such as small or flat lesions. Flat polyps can be detected with the support of CT [24, 25] although its detection supposes additional patient radiation and is limited by the size. Detection of small polyps cannot be undertaken with the help of CT as the current available resolution makes it impossible to detect polyps with size smaller than 10 mm as stated in [26], therefore the diagnosis in these cases should only

**•** Polyp classification: the decision of performing polypectomy is commonly taken by an estimation of the size and histology of the detected lesion. This estimation is commonly made by means of visual observation and therefore incorporates some degree of subjectivity. In this context, a system that can objectively provide an estimation of the size and classifi‐ cation of the polyp could allow taking in-vivo diagnostic decisions and this would optimize

**•** Patients lesion follow-up and endoscopy navigation: there is a necessity expressed by some clinicians regarding the recognition of the area that a lesion occupies, which can be useful for two different reasons: 1) for the case of polyps that have not been removed, an univocal recognition of the lesion would allow the study of the evolution of the lesion; 2) an accurate recognition of the marks that clinicians leave to identify the area of the polyp once it is removed would allow the exploration of areas nearby the lesion to search for new pathol‐

**3. Image processing challenges for the analysis of colonoscopy videos**

the performance of polyp localization methods, as shown in [13].

In order to provide clinicians with meaningful applications, the content of colonoscopy videos and frames must be thoroughly analyzed by computer scientists to search for lesions or indicators defined by clinicians. In this context, the majority of the literature has been focused on developing methods to characterize accurately the different elements of the endoluminal scene, paying special attention to polyps. Although it is clear that anatomical landmarks recognition is essential for application development, the acquisition and generation of high quality images is also crucial for computer vision methods in order to work as they are intended. For instance, the presence of image artifacts has been proven to have an impact in

Considering this we present in this section a summary of the most important challenges that a given computer vision method must face in order to provide with efficient support to clinicians. We have divided the challenges in two groups: those related to image acquisition

puter science may play a role have been identified:

rely on endoscopic exploration.

116 Screening for Colorectal Cancer with Colonoscopy

the treatment timing.

ogies.

**scientists**

Videos that endoscopes generate are created following common television standards in a way such they can provide with sufficiently moving image quality while allowing for efficient resource management in case endoscopy images and videos are stored for later inspection. It is important to mention that quality in this case is understood under human's observer point of view but not under computer visions; for instance there are some image processing techniques automatically performed – i.e. sharpening - that may improve how images are observed but, as they modify the original image, they create new elements that affect an automatic analysis by means of computer vision methods. Some of the features that can affect the performance of a computer vision method are listed below and in table 2: Videos that endoscopes generate are created following common television standards in such a way that they can provide sufficiently moving image quality while allowing for efficient resource management in case endoscopy images and videos are stored for later inspection. It is important to mention that quality in this case is understood under human's observer point of view but not under computer visions; for instance, there are some image processing techniques automatically performed – i.e., sharpening – that may improve how images are observed but, as they modify the original image, they create new elements that affect an automatic analysis by means of computer vision methods. Some of the features that can affect the performance of a computer vision method are listed below and in Table 2:

**•** *Illumination effects:* The way colonoscope illuminates the scene produces an axial illumina‐ tion which tends to generate *specular highlights* on shiny surfaces such as the mucosa. Mucosa is covered by a thin watery film which generates many specular highlights when it is illuminated in a perpendicular direction to its surface. Specular highlights position will vary with little movements of the colonoscope which will change the angle at which mucosa is illuminated therefore areas of the mucosa affected by specularities will change rapidly. The presence of specular highlights difficult strongly image processing [13] as they appear as very prominent structures which also hinder color and texture information about the surfaces in which they appear. Moreover, axial illumination introduces also an additional side-effect regarding its *lack of uniformity* in the way structures are illuminated: structures closer to the endoscope will appear brighter than others far from the endoscope (see Fig. 4). *a. Illumination effects:* The way colonoscope illuminates the scene produces an axial illumination which tends to generate *specular highlights* on shiny surfaces such as the mucosa. Mucosa is covered by a thin watery film which generates many specular highlights when it is illuminated in a perpendicular direction to its surface. Specular highlight positions will vary with little movements of the colonoscope which will change the angle at which mucosa is illuminated; therefore, areas of the mucosa affected by specularities will change rapidly. The presence of specular highlights makes image processing highly difficult (Bernal et al. (2015)) as they appear as very prominent structures which also hinder color and texture information about the surfaces in which they appear. Moreover, axial illumination introduces also an additional side effect regarding its *lack of uniformity* in the way structures are illuminated: structures closer to

the endoscope will appear brighter than others far from the endoscope (see Fig. 4).

Figure 4: Examples of illumination effects: (a) specular highlights, (b) overexposed polyp, and (c) underexposed polyp. Polyps in images b and c are delimited with a blue **Figure 4.** Examples of illumination effects: (a) specular highlights (b) overexposed polyp and (c) underexposed polyp. Polyps in images b and c are delimited with a blue mask to ease visualization.

mask to ease visualization. **•** *Sensor acquisition effects: Color phantoms* appear due to temporal misalignment of color channels related to some endoscopes that still use monochrome sensors. In this case, color information is generated by illuminating the scene with the three primary colors (red, green and blue) successively. Consequently, three different images are needed to generate a color image. This process introduces some undesired side-effects associated to camera movement: as we acquire the images in different time instants, specular highlights generated by the light source in each of the three moments will be located in slightly different positions, causing instability in the final color image –Fig. 5(a).

(e) (f) **Figure 5.** Effect of channel misalignment due to monochrome sensors: instability in specular highlights position (a) and apparition of color phantoms (b).

(c) (d)

Moreover, as each color channel is acquired in different times, the three components (red, green and blue) will not be exactly aligned if the endoscope moves when the image is acquired. This lack of color channel alignment generates artificial color bands in the contours of the structures –Fig. 5(b) - that appear in the image which limits the performance of any color informationbased structure characterization method.

**•** *Image resolution:* Commercial endoscopes generate videos in formats following television standards (PAL for Europe, NTSC for America and Japan). These formats are meant to generate motion images with enough quality to be observed by the general public but also minimizing the size of the information to be transmitted. By acting this way, videos generated by commercial endoscopes can be played in any standard system (TV, personal computers) without needing format conversion. Moreover, the minimization of the amount of transmitted information allows a reduction of the storage needs which is crucial in clinical settings where the amount of resources dedicated to information storage must be efficiently distributed.

Although the use of standard formats presents clear advantages for visualization and storage purposes, it does not benefit image processing by means of computer vision. Video standards offer images with lower resolution than the one that can be achieved by means of commercial cameras. For instance, NTSC standard provides as output 0.3 Megapixels images, HD standard offers images up to 2 Megapixels and a commercial camera easily exceeds 10 Megapixels [27]. Low resolution images lead to a loss of texture information associated to anatomical structures in the endoluminal scene, which can have an impact on the output of structure classification methods -Fig. 6-.

image. This process introduces some undesired side-effects associated to camera movement: as we acquire the images in different time instants, specular highlights generated by the light source in each of the three moments will be located in slightly different positions,

(a) (b)

(c) (d)

(e) (f) **Figure 5.** Effect of channel misalignment due to monochrome sensors: instability in specular highlights position (a) and

Moreover, as each color channel is acquired in different times, the three components (red, green and blue) will not be exactly aligned if the endoscope moves when the image is acquired. This lack of color channel alignment generates artificial color bands in the contours of the structures –Fig. 5(b) - that appear in the image which limits the performance of any color information-

**•** *Image resolution:* Commercial endoscopes generate videos in formats following television standards (PAL for Europe, NTSC for America and Japan). These formats are meant to generate motion images with enough quality to be observed by the general public but also minimizing the size of the information to be transmitted. By acting this way, videos generated by commercial endoscopes can be played in any standard system (TV, personal computers) without needing format conversion. Moreover, the minimization of the amount of transmitted information allows a reduction of the storage needs which is crucial in clinical settings where the amount of resources dedicated to information storage must be efficiently

Although the use of standard formats presents clear advantages for visualization and storage purposes, it does not benefit image processing by means of computer vision. Video standards offer images with lower resolution than the one that can be achieved by means of commercial cameras. For instance, NTSC standard provides as output 0.3 Megapixels images, HD standard offers images up to 2 Megapixels and a commercial camera easily exceeds 10 Megapixels [27]. Low resolution images lead to a loss of texture information associated to anatomical structures in the endoluminal scene, which can have an impact on the output of structure classification

causing instability in the final color image –Fig. 5(a).

118 Screening for Colorectal Cancer with Colonoscopy

apparition of color phantoms (b).

distributed.

methods -Fig. 6-.

based structure characterization method.

(c) (d) (e) (f) **Figure 6.** Different colonoscopy images acquired at different resolutions: (a) high resolution image and (b) low resolu‐ tion image. We can observe greater texture details in the polyp from the highest resolution image.

**•** *Image interlacing:* As it has been mentioned before, from all available video standards those with lowest bandwidth –amount of information that needs to be transmitted-requirements are chosen for use in endoscopy. This reduction in bandwidth is achieved by interlacing image lines, which is performed by acquiring odd and even image lines in different time instants. By this we can double the image refresh rate without increasing the size of the information. This also makes video movement appear smoother and more continuous to the human eye but it has a counterpart that affects posterior image processing. The final image provided by the processor will be a mixture of two different images captured in different time instants: even lines will be from the first capture whereas odd lines will come from the second. As with color channel misalignment, interlacing impact will depend on the amount of endoscope movement between the two acquisitions. For instance, if camera moves horizontally we can observe sawtooth profiles in vertical contours, apart from change of position of specular highlights. We show in Fig. 7 a clear example on how interlacing can affect the quality of the image to be processed by, for instance, the apparition of double and shadowy contours surrounding the elements of the image.

(c) (d) **Figure 7.** Impact of interlacing in image quality: (a) Interlaced image and (b) Separate field of an interlaced image.

(e) (f)

**•** *Sharpening:* Endoscopes and video processors include functionalities that improve the quality of the image to be visualized by human observers, aiming to simplify the observation of particular structures in the images. One of the most common techniques is sharpening, which describes a subjective perception of sharpness related to edge contrast in an image. By applying this technique, contours that separate different objects in the image can be more clearly identified and consequently structures can be easily separated –Fig. 8 (b)-. This visualization enhancement [28] comes at a cost in terms of image processing as contour enhancement implies a modification of the original image which increases image noise. Sharpening also generate halos around structures that appear in the image such as specular highlights, as observed in Fig. 8 (b).

(c) (d) **Figure 8.** Examples of sharpening applied on colonoscopy images: (a) Original image and (b) image with sharpening applied.


Building up the Future of Colonoscopy – A Synergy between Clinicians and Computer Scientists http://dx.doi.org/10.5772/61012 121

**Figure 9.** Examples of information overlay in colonoscopy images.

**•** *Sharpening:* Endoscopes and video processors include functionalities that improve the quality of the image to be visualized by human observers, aiming to simplify the observation of particular structures in the images. One of the most common techniques is sharpening, which describes a subjective perception of sharpness related to edge contrast in an image. By applying this technique, contours that separate different objects in the image can be more clearly identified and consequently structures can be easily separated –Fig. 8 (b)-. This visualization enhancement [28] comes at a cost in terms of image processing as contour enhancement implies a modification of the original image which increases image noise. Sharpening also generate halos around structures that appear in the image such as specular

(a) (b)

(c) (d)

**Figure 8.** Examples of sharpening applied on colonoscopy images: (a) Original image and (b) image with sharpening

(e) (f)

**•** *Information overlay:* Video processors associated to endoscope do not present a specific output dedicated to its connection to a personal computer. Considering this, the image that the clinician is observing will be the same that will be stored for later processing. It is common that some information regarding the procedure such as patient information or procedure date is superimposed to the image provided by the colonoscope, as it can be observed in Fig. 9. The presence of this information precludes its use for research purposes, as this data should be anonymzed. Moreover the presence of this information superimposed to the original image may difficult the observation and characterization of structures in the images apart from introducing additional noise and elements (letters, numbers) to the

**•** *Black mask:* Endoscopes automatically add an octagonal or circular black mask surrounding the image acquired by the sensor. This mask covers those regions of the image that are strongly affected by geometric distortions introduced by wide angle optic used in endo‐ scopes. These distortions, similar to fisheye effects present in some cameras, makes struc‐ tures below the mask appear different to what they are in reality and consequently they should not be analyzed by clinicians. Unfortunately the presence of this black mask affects the performance of image processing methods, as the mask creates strong contours in the separation between the mask and the endoluminal scene, as it can be observed in Fig. 10.

highlights, as observed in Fig. 8 (b).

120 Screening for Colorectal Cancer with Colonoscopy

applied.

image.

(c) (d) **Figure 10.** Impact of black mask in image processing algorithms. (a) shows the original image whereas (b) shows the output of an edge detection algorithm. Note that mask contours appear as strong as structural elements.

(e) (f) **•** *Data compression:* Image and video data are commonly compressed in order to save storage space but commonly used formats such as MPEG and JPEG lead to information loss along with the introduction of some artifacts they may difficult fine detail processing in images. In this case the lower the compression, the least impact it will have in further image processing.

#### **3.2. Endoluminal scene description challenges**

In order to provide with systems that can help clinicians to overcome some of the clinical challenges identified earlier, a description of the elements of the endoluminal scene is needed. We show in Fig.11 an example on how endoluminal scene looks like.

We can make a division of the elements that appear on a given scene into pure anatomical structures (polyps, luminal region, folds, blood vessels or intestinal content) and structures appearing as result of image acquisition and formation processes (specular highlights and black mask). It is clear that a potential intelligent system should focus on the characterization of anatomical structures in order to be clinically useful –being polyps the usual target structurebut, as recent studies demonstrate [29], the consideration of all the elements of the endoluminal scene may result in an improvement of the performance of a given system. Endoluminal structure characterization is not a straightforward task due to three main reasons:

**Figure 11.** Elements of the endoluminal scene: (1) Polyp; (2) Luminal region; (3) Folds; (4) Blood vessels; (5) Intestinal content; (6) Specular highlights and (7) Black mask.

**•** *Lack of uniform structure appearance:* Anatomical structures appearance differs greatly in different interventions, which may difficult the development of characterization methods that can be widely applicable. For instance, polyp characterization is challenging because there is not an uniform and unique polyp appearance; in fact, polyp appearance depends greatly on the point of view in which it is observed and we can observe different particu‐ larities whether we are observing polyps in zenithal or lateral views –see Fig. 12-.

(c) (d) **Figure 12.** Variability in polyp appearance: (1) Zenithal view and (2) Lateral view.

We can make a division of the elements that appear on a given scene into pure anatomical structures (polyps, luminal region, folds, blood vessels or intestinal content) and structures appearing as result of image acquisition and formation processes (specular highlights and black mask). It is clear that a potential intelligent system should focus on the characterization of anatomical structures in order to be clinically useful –being polyps the usual target structurebut, as recent studies demonstrate [29], the consideration of all the elements of the endoluminal scene may result in an improvement of the performance of a given system. Endoluminal

**Figure 11.** Elements of the endoluminal scene: (1) Polyp; (2) Luminal region; (3) Folds; (4) Blood vessels; (5) Intestinal

**•** *Lack of uniform structure appearance:* Anatomical structures appearance differs greatly in different interventions, which may difficult the development of characterization methods that can be widely applicable. For instance, polyp characterization is challenging because there is not an uniform and unique polyp appearance; in fact, polyp appearance depends greatly on the point of view in which it is observed and we can observe different particu‐

larities whether we are observing polyps in zenithal or lateral views –see Fig. 12-.

content; (6) Specular highlights and (7) Black mask.

122 Screening for Colorectal Cancer with Colonoscopy

structure characterization is not a straightforward task due to three main reasons:

(e) (f) Consequently a definition of a model of appearance for a given structure should consider this great variability in order to be widely applicable and, therefore, search for general features that can be attainable for the majority of the cases. (a) (b) Figure 12: Variability in polyp appearance: (1) zenithal view and (2) lateral view.

**•** *Impact of other elements of the scene on a particular element characterization:* Following with the polyp example, the majority of available works rely on polyp characterization from the identification of polyp boundaries but, in terms of image processing, there is not a big difference in terms of contour appearance between polyps, blood vessels and folds, as the three of them provide with similar response to contour detection operators, as it can be observed in Fig. 13. Considering this, a given intelligent system must consider the impact of all present structures when providing a characterization of a particular one and it will need to find additional cues to differentiate between these structures. Consequently, a definition of a model of appearance for a given structure should consider this great variability in order to be widely applicable and, therefore, search for general features that can be attainable for the majority of the cases. b. *Impact of other elements of the scene on a particular element characterization:* Following with the polyp example, the majority of available works rely on polyp characterization from the identification of polyp boundaries but, in terms of image processing, there is not a big difference in terms of contour appearance between polyps, blood vessels, and folds, as the three of them provide with similar response to contour detection operators, as observed in Fig. 13. Considering this, a given intelligent system must consider the

> impact of all present structures when providing a characterization of a particular one and it will need to find additional cues to differentiate between these structures.

Number 1 represents a polyp, number 2 a fold, and 3 represents blood vessels. c. *Difficulties in the definition of the structural element:* Another challenge is related to the **Figure 13.** Example of similarity of response of different structures to a given operator. Number 1 represents a polyp, number 2 a fold and 3 represents blood vessels.

visual definition of the structure itself, that is, sometimes the definition of the element itself is not clear, which makes it difficult to delimit the structure. For instance, recent studies show a great variability between observers when defining the luminal region – demonstrated in Sánchez et al. (2015) – which may have an impact on ground truth creation for assessing the performance of a given intelligent system. This difficulty in **•** *Difficulties on the definition of the structural element:* Another challenge is related to the visual definition of the structure itself, that is, sometimes the definition of the element itself is not clear, which makes it difficult to delimit the structure. For instance, recent studies show a great variability between observers when defining the luminal region –demonstrated in [30], which may have an impact on ground truth creation for assessing the performance of a given intelligent system. This difficult on the definition on the structure can also be applied for other elements such as fecal or intestinal content.

## **4. Equipment setting to favor optimal image processing analysis**

We present in this section the optimal settings of clinical equipment to ensure the best possible quality of the images which will be analyzed by the intelligent system.

#### **4.1. Endoscopic equipment settings**

Chronologically, the first element to be considered is the configuration of both endoscope and video processor in order to obtain the best possible images for further analysis. In this case we propose the following configuration:


#### **4.2. Image storage and anonymization**

We have to consider that image or/and video data will be used in research projects from which several research publications will be generated. Access to this image or video data should be granted to other researchers in order to allow an easier comparison of the performance of different methods. Considering this, no information that can allow an identification of either the patient or the clinician should be provided in neither the images or in the metadata associated to them –such as time and date of image capture or endoscopy used-, preventing the association of a given image to a patient, clinician or hospital.

Considering the amount of endoscopic interventions performed in a hospital in a year, images or videos that are stored tend to be compressed. This compression has already been mentioned to have implications for image processing methods so; if possible, the configuration with less possible compression should be chosen.

#### **4.3. Endoscopic naviagation guidelines**

Endoscope movement when images are acquired impacts the quality of the images that are obtained. If there is no scope movement, effects such as interlacing or color phantoms can be almost inexistent -Fig. 14 (a)-. Considering this, we propose still images acquisition to be made being both the scope and the elements of the endoluminal scene static. For the case of video acquisition we suggest slow and smooth endoscope progression through the patient in order to maximize the reduction of movement-related artifacts generation.

**4. Equipment setting to favor optimal image processing analysis**

quality of the images which will be analyzed by the intelligent system.

**4.1. Endoscopic equipment settings**

124 Screening for Colorectal Cancer with Colonoscopy

propose the following configuration:

**4.2. Image storage and anonymization**

possible compression should be chosen.

**4.3. Endoscopic naviagation guidelines**

We present in this section the optimal settings of clinical equipment to ensure the best possible

Chronologically, the first element to be considered is the configuration of both endoscope and video processor in order to obtain the best possible images for further analysis. In this case we

**•** Disable sharpening options, so we can avoid the apparition of artificial information (halos)

**•** Disable the superimposition of overlay information such as patient or procedure data to obtain a clean view of the endoluminal scene. This also allows a complete anonymization

**•** If possible, allow the endoluminal view to occupy the largest portion of the scene without

We have to consider that image or/and video data will be used in research projects from which several research publications will be generated. Access to this image or video data should be granted to other researchers in order to allow an easier comparison of the performance of different methods. Considering this, no information that can allow an identification of either the patient or the clinician should be provided in neither the images or in the metadata associated to them –such as time and date of image capture or endoscopy used-, preventing

Considering the amount of endoscopic interventions performed in a hospital in a year, images or videos that are stored tend to be compressed. This compression has already been mentioned to have implications for image processing methods so; if possible, the configuration with less

Endoscope movement when images are acquired impacts the quality of the images that are obtained. If there is no scope movement, effects such as interlacing or color phantoms can be almost inexistent -Fig. 14 (a)-. Considering this, we propose still images acquisition to be made being both the scope and the elements of the endoluminal scene static. For the case of video acquisition we suggest slow and smooth endoscope progression through the patient in order

**•** Configure storage options to obtain data with the minimum possible compression.

surrounding image structure contours along with reducing image noise.

of the information easing its use for research purposes.

the association of a given image to a patient, clinician or hospital.

to maximize the reduction of movement-related artifacts generation.

applying any kind of digital zooming operation.

(c) (d) (e) (f) **Figure 14.** Difference in image quality related to endoscope movement when acquiring images: (a) still endoscope vs. (b) moving endoscope.

It is clear that even by considering all the suggestions expressed, there will still be a minor movement of the scope between the two time instants in which odd and even lines of the final image are acquired. In order to mitigate the impact of interlacing and to avoid loss of image resolution we propose to make a real-time analysis of the images when they are acquired in order to store only the one which less interlacing impact. This analysis will be made by comparing consecutive frames, where the difference in content between them is so minimal that there is no point on storing them all, considering the small changes that will appear in images extracted from a 30 frames per second video. In case interlacing can still be perceived, its impact can be completely removed by working with one of the two channels of the image [29], although this implies a decrease in final image resolution.

To close this section, we show in Table 2 a summary of the challenges related to image formation and acquisition depicted in Section 3 and our proposal on how to solve/mitigate them. As it can be seen from the table, there are some challenges that cannot be solved by applying specific settings to the devices involved. For instance, those related to image formation are highly device-dependent. In this sense, newer equipment has dedicated sensors for each color channel avoiding the apparition of color phantoms. There are other challenges that must be solved by means of image processing techniques, such as specular highlights. In this sense, the most accepted solution [29] consists of a specular highlight detection followed by a substitution of the pixels in the image belonging to specular highlights by a combination of valid values of neighbor pixels, as it can be observed in Fig. 15. The same operation is applied to mitigate the impact of strong contours created by the black mask.

(c) (d) (e) (f) **Figure 15.** Application of image processing methods to mitigate impact of specular highlights and black mask. (a) Original image and (b) Processed image.


**Table 2.** Summary of image acquisition and formation challenges along with proposal of solutions

#### **5. Current endoluminal scene description methods**

We present in this section a review on the most recent works published on the topic of anatomical endoluminal scene elements description.

#### **5.1. Polyps**

As they are the main focus of colonoscopy explorations, the majority of already existing intelligent systems for colonoscopy deals with polyp characterization. We divide existing systems according to the application they are built for:

**•** *Polyp detection:* This group of methods aim to decide whether there is a polyp or not in the image. The majority of the works on polyp detection are built on the principle of applying a given feature detector/descriptor to the image in order to guide detection methods. In this sense, we can divide existing approaches in two groups: (a) shape and (b) texture and colorbased. The first group aims to detect polyps by observing specific cues on the contours of the polyp –examples of this can be found in works presented in [31-33], or by fitting candidate objects in the image to the most common shapes that polyps present [34]. Regarding the second group, the use of several general descriptors has been proposed, such as wavelets in [35], local binary patterns in [36] or co-ocurrence matrices [37]. A method combining MPEG-7 texture and color descriptors was proposed in [38]. One big drawback of descriptor-based methods is that they tend to need of an exhaustive training and they are

**Figure 16.** Example of the output of each polyp characterization group of algorithms.

**Source Challenge Proposed solution**

Image acquisition and

Image visualization capabilities enhancement

storage

126 Screening for Colorectal Cancer with Colonoscopy

Image acquisition and visualization

**5.1. Polyps**

Image formation Illumination Specular highlights Specular highlights correction

Sensor acquisition Color phantoms Device-dependent

Presence of patient and procedure

We present in this section a review on the most recent works published on the topic of

As they are the main focus of colonoscopy explorations, the majority of already existing intelligent systems for colonoscopy deals with polyp characterization. We divide existing

**•** *Polyp detection:* This group of methods aim to decide whether there is a polyp or not in the image. The majority of the works on polyp detection are built on the principle of applying a given feature detector/descriptor to the image in order to guide detection methods. In this sense, we can divide existing approaches in two groups: (a) shape and (b) texture and colorbased. The first group aims to detect polyps by observing specific cues on the contours of the polyp –examples of this can be found in works presented in [31-33], or by fitting candidate objects in the image to the most common shapes that polyps present [34]. Regarding the second group, the use of several general descriptors has been proposed, such as wavelets in [35], local binary patterns in [36] or co-ocurrence matrices [37]. A method combining MPEG-7 texture and color descriptors was proposed in [38]. One big drawback of descriptor-based methods is that they tend to need of an exhaustive training and they are

information

**Table 2.** Summary of image acquisition and formation challenges along with proposal of solutions

**5. Current endoluminal scene description methods**

anatomical endoluminal scene elements description.

systems according to the application they are built for:

Lack of uniform illumination Device-dependent

Sharpening Disable sharpening

Black mask Black mask substitution Data compression Use of lossless compression

Image resolution Stabilization of endoscope,

Image interlacing Interlacing suppression, neighbor

interlacing suppression and use of

HD endoscopes

stabilization

standards.

Disable overlays

frame frames, endoscope

very sensitive to parameter tuning. Finally the work published in [39] combines shape and texture features to build up a polyp detection method which also considers spatial and temporal adjacency information present in colonoscopy videos.


As it can be seen from the classification exposed above, a potential intelligent system with applicability in the intervention room could easily use a system from each of the four groups in order to build up a computer-aided diagnosis tool. We show in Fig. 16 a graphical example of such a system. In a first stage the system will automatically decide which frames contain a polyp and which region of the frame contains the polyp. From this, an accurate segmentation of the polyp region will be obtained in order to extract meaningful features to help in the classification process.

#### **5.2. Luminal area**

Luminal area is defined as the interior space of a tubular structure, such as the intestine. The detection of the lumen and its position can be crucial in both intervention and post-intervention time.

On the one hand, an accurate detection of the lumen region during in-vivo intervention may be useful to discard areas of the image with low visibility –Fig. 17(a) - in order to save computation time for other interesting regions of the image as proposed in [44]. Lumen detection can also be helpful to guide the clinician inside the intestine by pointing out which direction he/she should take to progress. On the other hand, lumen characterization in postintervention can be used to discard frames for further revision: frames where the proportion of lumen out of the entire image is large can be related to the progression of the colonoscope through the gut but, conversely, frames where the amount of lumen presence is low may potentially indicate areas of the image where the physician has paid more attention. This can be useful to obtain summary videos of the whole procedure. Lumen characterization has been an active topic of research in several endoscopy image modalities such as optical –works of [45] and [46] - and virtual colonoscopy [47]. The main reasoning behind the majority of the luminal region characterization methods is the assumption that lumen is the darkest region of the image and from this seed region growing algorithms are built in order to find lumen boundaries.

#### **5.3. Blood vessels**

Blood vessels are the part of the circulatory system that transports blood through the body and they can be identified by their tree-like shape with ramifications. The characterization of these branching structures has been reported in domains such as retinal image analysis [48] or palm prints recognition [49]. Blood vessels characterization in colonoscopy images can be useful in two domains: helping in polyp localization and segmentation tasks, as it has been proven in [13, 29, 42], and as key points to be used in potential follow-up methods, as proposed in [50]. Regarding the former, a mitigation of blood vessels related valleys by using contrast properties of blood vessels contours has been proven to be useful to improve polyp localization segmen‐ tation, as in some images -Fig. 17(b)- blood vessels can be identified easier than polyp boun‐ daries. Concerning the latter, we could think of a univocal characterization of blood vessels branching patterns using methods such as the one proposed in [51] to recognize a same region during different interventions.

#### *5.3.1. Folds*

in order to build up a computer-aided diagnosis tool. We show in Fig. 16 a graphical example of such a system. In a first stage the system will automatically decide which frames contain a polyp and which region of the frame contains the polyp. From this, an accurate segmentation of the polyp region will be obtained in order to extract meaningful features to help in the

Luminal area is defined as the interior space of a tubular structure, such as the intestine. The detection of the lumen and its position can be crucial in both intervention and post-intervention

On the one hand, an accurate detection of the lumen region during in-vivo intervention may be useful to discard areas of the image with low visibility –Fig. 17(a) - in order to save computation time for other interesting regions of the image as proposed in [44]. Lumen detection can also be helpful to guide the clinician inside the intestine by pointing out which direction he/she should take to progress. On the other hand, lumen characterization in postintervention can be used to discard frames for further revision: frames where the proportion of lumen out of the entire image is large can be related to the progression of the colonoscope through the gut but, conversely, frames where the amount of lumen presence is low may potentially indicate areas of the image where the physician has paid more attention. This can be useful to obtain summary videos of the whole procedure. Lumen characterization has been an active topic of research in several endoscopy image modalities such as optical –works of [45] and [46] - and virtual colonoscopy [47]. The main reasoning behind the majority of the luminal region characterization methods is the assumption that lumen is the darkest region of the image and from this seed region growing algorithms are built in order to find lumen

Blood vessels are the part of the circulatory system that transports blood through the body and they can be identified by their tree-like shape with ramifications. The characterization of these branching structures has been reported in domains such as retinal image analysis [48] or palm prints recognition [49]. Blood vessels characterization in colonoscopy images can be useful in two domains: helping in polyp localization and segmentation tasks, as it has been proven in [13, 29, 42], and as key points to be used in potential follow-up methods, as proposed in [50]. Regarding the former, a mitigation of blood vessels related valleys by using contrast properties of blood vessels contours has been proven to be useful to improve polyp localization segmen‐ tation, as in some images -Fig. 17(b)- blood vessels can be identified easier than polyp boun‐ daries. Concerning the latter, we could think of a univocal characterization of blood vessels branching patterns using methods such as the one proposed in [51] to recognize a same region

classification process.

128 Screening for Colorectal Cancer with Colonoscopy

**5.2. Luminal area**

time.

boundaries.

**5.3. Blood vessels**

during different interventions.

Haustral folds represent folds of mucosa within the colon. They are formed by circumferential contraction of the inner muscular layer of the colon. In the context of intelligent systems for colonoscopy, folds characterization can play a key role in polyp characterization tasks. In this sense, we have to consider that the fold contours appearance in colonoscopy images is very similar to the one of polyps. We can observe in Fig. 17 (c) that folds and polyp contours present similar appearance but different levels of curvature; consequently, an accurate identification of folds could lead to an improvement in polyp characterization tasks. Some recent works build up advances model of polyp appearance to discriminate polyp contours from folds by considering desirable properties of polyp contours such as concavity, completeness or continuity, as proposed in [13].

**Figure 17.** Effect of endoluminal scene structures in polyp characterization: (a) Luminal region (delimited by a blue mask); (b) Blood vessels and (c) Folds.

#### **5.4. Fecal content**

Apart from the elements that have already been covered, there are more elements that can appear in the endoluminal scene as a result of bad patient preparation. In this sense high presence of intestinal content is considered by clinicians as an indicator to decide whether a procedure has to be repeated or not as no clinician or computer vision method would work with very low quality images. Moreover, there are some cases when the presence of fecal content can affect the output of computer vision methods, as it was shown in [13]. Therefore an accurate identification of fecal content in colonoscopy images could be used to provide automatic indicators of the quality of patients' preparation.

#### **6. Building up validation frameworks for intelligent systems**

One of the main problems when assessing the performance of the different available intelligent systems for colonoscopy is that the majority of them are tested on private databases, which makes it difficult to observe the differences in performance between them and to extrapolate its functioning in other environments. Moreover, it is very difficult to compare performance levels of different methods as each of them proposes or uses different evaluation metrics which, for some cases, can be only used with a specific application in mind. Considering this two problems, we present in this section our proposal for a complete validation framework covering from database and ground truth creation to the definition of the metrics to be used to evaluate a given method.

#### **6.1. Database creation**

In order to validate and assess the performance of a computer vision method, this has to be tested in a set of images covering as many possible cases of study. For instance, if we want our method to be able to characterize polyps from all the types present in Paris classification our database should contain several examples from each of the classes that are defined there. Apart from the original images, a ground truth should also be provided. This ground truth will be used to assess the performance of the method and its configuration will depend on the concrete experiment. Following the same example used before, for polyp localization purposes the ground truth should consist of a binary image where pixels in white should correspond to those pixels which are part of the polyp. If the output of a given method falls in the white pixels of the image, the method will be performing as expected. As it can be seen there are two processes involved when creating databases for intelligent systems validation: the selection of the cases to be included in the database and the creation of the corresponding ground truth.

Regarding the selection of the cases, in order the use of a method can be extended outside research domain, these cases should represent the clinical variability that the clinician can find during interventions. In case we have several types of elements to be characterized, the database should contain as many different examples as possible for all the possible classes. It is important to mention than the more different the examples, the more robust will be our method and the better it will perform once a new case of study is to be analyzed. By doing this if we achieve that a given method offers good performance in our database it will be easy to extrapolate its performance in a potential clinical application.

There is one branch of computer vision known as machine learning which involves method training in a set of images and a posterior testing of this method in a different set of images, once its performance has been optimized in the training stage. Considering this, the size of the database should permit the division in training and testing examples and we should define our database in a way such representative examples of all the possible cases are present both in training and testing databases. The final size of the database should allow extracting statistically significant conclusions. In clinical trials, a variability of less of 10 % is not consid‐ ered as relevant as stated in [52], being variability calculated as the inverse of the square root of the number of samples –N- in our database. Considering this, the minimum size of the database should be of 100 images.

Once database has been defined, ground truth must be created to validate the performance of the methods. The definition of this ground truth is clearly application dependent: for instance if we are developing a polyp detection method the ground truth may only consist of an excel file indicating for each frame whether there is a polyp or not in the image but for a polyp segmentation method we would need a binary image representing the structure to be seg‐ mented, as it can be seen in Fig. 18. may only consist of an excel file indicating for each frame whether there is a polyp or not in the image, but for a polyp segmentation method we would need a binary image

dependent: for instance, if we are developing a polyp detection method, the ground truth

√��. Considering this, the minimum size of the database should be of <sup>100</sup> images.

representing the structure to be segmented, as seen in Fig. 18.

Regarding the selection of the cases, so that the use of a method can be extended outside research domain, these cases should represent the clinical variability that the clinician can find during interventions. In case we have several types of elements to be characterized, the database should contain as many different examples as possible for all the possible classes. It is important to mention that the more different the examples, the more robust will be our method and the better it will perform once a new case study is to be analyzed. By doing this, if a given method offers good performance in our database it will be easy to extrapolate its

There is one branch of computer vision known as machine learning which involves method training in a set of images and a posterior testing of this method in a different set of images, once its performance has been optimized in the training stage. Considering this, the size of the database should permit the division in training and testing examples and we should define our database in a way such that representative examples of all the possible cases are present both in training and testing databases. The final size of the database should allow extracting statistically significant conclusions. In clinical trials, a variability of less than 10% is not considered as relevant as stated in Julious et al. (2009), variability being calculated as the inverse of the square root of the number of samples – N – in our database (����������� �

performance in a potential clinical application.

�

Figure 18: Possible contents of a polyp segmentation database: (a) Original image; **Figure 18.** Possible contents of a polyp segmentation database: (a) Original image; (b) Polyp mask; (c) Polyp contour mask and (d) Black mask.

(b) Polyp mask; (c) Polyp contour mask; and (d) Black mask. Image‐based ground truth are commonly created using image editing software such as Microsoft Paint or Adobe Photoshop, although there is an increasing use of specific tools such as ImageJ (Abràmoff et al. (2004)), which allows creation of segmentation ground truths by marking a few points in the image. Concerning ground truth creation, it should be created either by clinicians or by experts under clinicians' supervision. Having more than Image-based ground truth are commonly created using image editing software such as Microsoft Paint or Adobe Photoshop, although there is an increasing use of specific tools such as ImageJ [53] which allows the creation of segmentation ground truths by marking a few points in the image. Concerning ground truth creation, it should be created either by clinicians or by experts under clinicians' supervision. Having more than one ground truth per image is recommendable for validation purposes as a way to avoid possible subjectivity in ground truth creation. This allows performing statistical tests and also to assess whether the performance of a given method is within inter-observer variability. If clinical conclusions are meant to be extracted from the performance of intelligent systems, clinical metadata should be provided. For instance, if we want to assess the performance of a polyp classification method, apart from the mask representing where the polyp in the image is, clinicians should provide which is the class of the polyp (i.e., KUDO type I).

Currently there are only, up to our knowledge, three different databases related to colonoscopy image analysis: two of them consisting of still images showing a polyp - CVC-ColonDB and CVC-ClinicDB- and another - ASU-Mayo Clinic polyp database-, which consists of full colonoscopy videos with and without polyps. The first two databases are meant for the validation of model of appearance for polyps to ease polyp localization and segmentation whereas the latter has been developed for the validation of polyp detection algorithms. Currently only CVC-ClinicDB incorporates clinical metadata associated to each polyp, including information regarding polyp size, Paris classification and histological type of polyp. This allows break down of the results according clinical criteria, as exposed in [13]. We introduce the main features of each of the three databases in Table 3.

#### **6.2. Performance metrics**

levels of different methods as each of them proposes or uses different evaluation metrics which, for some cases, can be only used with a specific application in mind. Considering this two problems, we present in this section our proposal for a complete validation framework covering from database and ground truth creation to the definition of the metrics to be used

In order to validate and assess the performance of a computer vision method, this has to be tested in a set of images covering as many possible cases of study. For instance, if we want our method to be able to characterize polyps from all the types present in Paris classification our database should contain several examples from each of the classes that are defined there. Apart from the original images, a ground truth should also be provided. This ground truth will be used to assess the performance of the method and its configuration will depend on the concrete experiment. Following the same example used before, for polyp localization purposes the ground truth should consist of a binary image where pixels in white should correspond to those pixels which are part of the polyp. If the output of a given method falls in the white pixels of the image, the method will be performing as expected. As it can be seen there are two processes involved when creating databases for intelligent systems validation: the selection of the cases to be included in the database and the creation of the corresponding ground truth.

Regarding the selection of the cases, in order the use of a method can be extended outside research domain, these cases should represent the clinical variability that the clinician can find during interventions. In case we have several types of elements to be characterized, the database should contain as many different examples as possible for all the possible classes. It is important to mention than the more different the examples, the more robust will be our method and the better it will perform once a new case of study is to be analyzed. By doing this if we achieve that a given method offers good performance in our database it will be easy to

There is one branch of computer vision known as machine learning which involves method training in a set of images and a posterior testing of this method in a different set of images, once its performance has been optimized in the training stage. Considering this, the size of the database should permit the division in training and testing examples and we should define our database in a way such representative examples of all the possible cases are present both in training and testing databases. The final size of the database should allow extracting statistically significant conclusions. In clinical trials, a variability of less of 10 % is not consid‐ ered as relevant as stated in [52], being variability calculated as the inverse of the square root of the number of samples –N- in our database. Considering this, the minimum size of the

Once database has been defined, ground truth must be created to validate the performance of the methods. The definition of this ground truth is clearly application dependent: for instance if we are developing a polyp detection method the ground truth may only consist of an excel file indicating for each frame whether there is a polyp or not in the image but for a polyp

extrapolate its performance in a potential clinical application.

database should be of 100 images.

to evaluate a given method.

130 Screening for Colorectal Cancer with Colonoscopy

**6.1. Database creation**

The way a given intelligent system method is validated will depend greatly on what this intelligent system is for. The potential application the system is designed for will define both how database and ground truth need to be generated and the metrics used to assess the performance of the method. In this subsection we propose validation protocols for each of the four main types of intelligent systems reported in the literature.


**Table 3.** Summary of available databases for colonoscopy image analysis

**•** *Polyp Detection:* A given polyp detection method should provide an output whenever a polyp is present in the image and should not provide any output if there is no polyp.

Performance metrics:

Considering this we propose the use of four different concepts (True Positive (TP), False Positive (FP), True Negative (TN) and False Negative (FN)) which are commonly used in object detection and characterization problems. We present these concepts in Table 4.


**Table 4.** Explanation of polyp detection metrics

Consequently a good polyp detection method should provide with a high number of TP and TN along the lowest possible number of FN and FP. In order to allow a more clear represen‐ tation of these results, four different metrics are calculated from TP, FP, TN and FN values:

**•** Precision, calculated as: *Prec* <sup>=</sup> *TP TP* <sup>+</sup> *FP* . It represents the fraction of relevant retrieved information. Regarding polyp detection, it represents the percentage of correct alarms (frames where the method provides an output and the image has a polyp). A low precision rate will be interpreted as the system providing a high number of false alarms.


Finally, a polyp detection method will be considered as clinically useful if it can helps the clinician to detect the polyp. Considering this and assuming that a given sequence contains a polyp, the following metrics can be defined:


Considering this two metrics, a comparison can be made between the performance of a given automatic method and clinicians, as it was presented in [13]. This can allow the assessment of the potential of a given method to be included to support clinicians in polyp detection tasks.

#### *Ground truth*:

**Database Number of frames/videos Ground truth content**

4) polyp contour.

biopsy) [13]

polyp region is provided.

**•** *Polyp Detection:* A given polyp detection method should provide an output whenever a polyp is present in the image and should not provide any output if there is no polyp.

Considering this we propose the use of four different concepts (True Positive (TP), False Positive (FP), True Negative (TN) and False Negative (FN)) which are commonly used in object

Consequently a good polyp detection method should provide with a high number of TP and TN along the lowest possible number of FN and FP. In order to allow a more clear represen‐ tation of these results, four different metrics are calculated from TP, FP, TN and FN values:

information. Regarding polyp detection, it represents the percentage of correct alarms (frames where the method provides an output and the image has a polyp). A low precision

rate will be interpreted as the system providing a high number of false alarms.

detection and characterization problems. We present these concepts in Table 4.

**TP** Provides an output There is a polyp in the image

**FP** Provides an output There is no polyp in the image **TN** Does not provide an output There is no polyp in the image

**FN** Does not provide an output There is a polyp in the image

**Concept Method Ground truth**

For each image with a polyp the following images are provided: 1) original image; 2) polyp mask; 3) non-informative regions mask and

For each image both the original frame along with a mask covering the polyp are provided. For each polyp, clinical metadata associated is provided (size, Paris classification, histological type of polyp after

For each frame of the video a binary image is provided. Absence of polyp in the image can be identified by having a completely black associated image. In case of polyp presence, an approximation of

*TP* <sup>+</sup> *FP* . It represents the fraction of relevant retrieved

**CVC-ColonDB** 380 frames from 15 different

132 Screening for Colorectal Cancer with Colonoscopy

**CVC-ClinicDB** 612 frames from 29 different

**ASU-Mayo Clinic polyp database**

Performance metrics:

**Table 4.** Explanation of polyp detection metrics

**•** Precision, calculated as: *Prec* <sup>=</sup> *TP*

sequences with a polyp

sequences with a polyp.

Training set: 20 videos (10 with a polyp and 10 without polyps). Testing set: 18 videos

**Table 3.** Summary of available databases for colonoscopy image analysis

Ground truth for polyp detection methods validation can consist in either a text file stating which frames contain a polyp or in a binary mask corresponding to each original frame. In this case the binary mask should represent polyp presence and absence (for instance, an all-black image can represent polyp absence).

**•** *Polyp localization:* Polyp localization methods aim to extend the information provided by polyp detection methods by not only indicating whether there is a polyp in the image or not, but also indicating where the polyp is within the image.

#### *Performance metrics*:

Considering the purpose of localization methods, we cannot use all the four concepts explained before as the use of TN does not make sense in this type of problems as there is always a polyp in the image. In this case several authors [13] propose a more direct performance referred as localization accuracy. Considering that a polyp localization method always provide a potential polyp location, we can define a good localization (GL) whenever the output of the localization method coincides with a polyp. Conversely we define false localization (FL) in the opposite case when the localization proposed by the method falls outside the polyp. Taking this into account, we define localization accuracy as:

$$LAcc = \frac{GL}{GL + FL}$$

In cases where the output of a localization image does not consists of points representing polyp locations but of energy images representing areas with more likelihood of containing a polyp –as it can be seen in Fig. 16- the use of energy concentration metrics seems useful to represent the performance of a method [13]. Considering these two metrics, LAcc and concentration, a good localization method should provide a low number of FL while concentrating the majority of the polyp presence likelihood image inside the polyp mask. (Bernal et al. (2015)). Considering these two metrics, LAcc and concentration, a good localization method should provide a low number of FL while concentrating the majority of the polyp presence likelihood image inside the polyp mask. *Ground truth:* Ground truth for polyp localization should consist of binary masks representing the area of the image that is occupied by the polyp, as shown in Figure 18.

#### *Ground truth*:

Ground truth for polyp localization should consist of binary masks representing the area of the image that is occupied by the polyp, as it is shown in Figure 18. c. *Polyp segmentation:* An accurate segmentation of the region that contains the polyp can be useful for both lesion recognition tasks as well as for delimiting the area of the image to be used for lesion classification purposes.

**•** *Polyp segmentation:* An accurate segmentation of the region that contains the polyp can be useful for both lesion recognition tasks as well as for delimiting the area of the image to be used for lesion classification purposes. *Performance metrics:* We propose the use of common segmentation metrics such as Precision and Recall, as they were defined for polyp detection. In this case, we classify each pixel as TP,

FP, TN, and FN considering methods' output and the ground truth (i.e., a false

#### *Performance metrics*: positive pixel is defined as a pixel in which our method states it is part of the polyp

We propose the use of common segmentation metrics such as Precision andRecall, as they were defined for polyp detection. In this case we classify each pixel as TP, FP, TN and FN consider‐ ingmethods' output andthe groundtruth (i.e. a falsepositivepixel isdefinedas apixel in which our method states it is part of the polyp when it is not). In this context, a good polyp segmenta‐ tion method should provide higher Precision and Recall results (Fig. 19 (b)); a method provid‐ ing high Precision with low Recall will provide regions that cannot be used for further polyp characterizationas theycontainlotsofnon-polypinformation(Fig. 19 (c)).Converselyamethod providing with high Recall but low Precision values will be useful for polyp description but will leave a lot of useful polyp content out of posterior analysis (Fig. 19 (d)). when it is not). In this context, a good polyp segmentation method should provide higher Precision and Recall results (Fig. 19 (b)); a method providing high Precision with low Recall will provide regions that cannot be used for further polyp characterization as they contain lots of non‐polyp information (Fig. 19 (c)). Conversely, a method providing high Recall but low Precision values will be useful for polyp description but will leave a lot of useful polyp content out of posterior analysis (Fig. 19 (d)).

results with (a) good Precision and Recall values; (c) good Precision but low Recall value; and (d) low Precision but good Recall value. Mask representing the output of a given method is represented in blue. **Figure 19.** Interpretation of segmentations: (a) Original ground truth. Segmentation results with (a) good Precision and Recall values; (c) good Precision but low Recall value and (d) low Precision but good Recall value. Mask representing the output of a given method is represented in blue.

Figure 19: Interpretation of segmentations: (a) original ground truth. Segmentation

#### *Ground truth*:

*Ground truth:*

As for the case of polyp localization, ground truth for polyp segmentation should consists of binary masks representing either the area of the image that is occupied by the polyp -Figure 18 (b)- or the contour of the polyp region -Figure 18 (c)-.

**•** *Polyp classification:* A good polyp classification method should be able to assign the polyp present in the image the same label/class that is attached to the polyp in the ground truth.

#### *Performance metrics*:

<sup>=</sup> <sup>+</sup> *GL LAcc*

In cases where the output of a localization image does not consists of points representing polyp locations but of energy images representing areas with more likelihood of containing a polyp –as it can be seen in Fig. 16- the use of energy concentration metrics seems useful to represent the performance of a method [13]. Considering these two metrics, LAcc and concentration, a good localization method should provide a low number of FL while concentrating the majority

Ground truth for polyp localization should consist of binary masks representing the area of

c. *Polyp segmentation:* An accurate segmentation of the region that contains the polyp can be useful for both lesion recognition tasks as well as for delimiting the area of

the area of the image that is occupied by the polyp, as shown in Figure 18.

**•** *Polyp segmentation:* An accurate segmentation of the region that contains the polyp can be useful for both lesion recognition tasks as well as for delimiting the area of the image to be

We propose the use of common segmentation metrics such as Precision and Recall, as they were defined for polyp detection. In this case, we classify each pixel as TP, FP, TN, and FN considering methods' output and the ground truth (i.e., a false positive pixel is defined as a pixel in which our method states it is part of the polyp when it is not). In this context, a good polyp segmentation method should provide higher Precision and Recall results (Fig. 19 (b)); a method providing high Precision with low Recall will provide regions that cannot be used for further polyp characterization as they contain lots of non‐polyp information (Fig. 19 (c)). Conversely, a method providing high Recall but low Precision values will be useful for polyp description but will leave a lot of useful polyp content out of posterior

We propose the use of common segmentation metrics such as Precision andRecall, as they were defined for polyp detection. In this case we classify each pixel as TP, FP, TN and FN consider‐ ingmethods' output andthe groundtruth (i.e. a falsepositivepixel isdefinedas apixel in which our method states it is part of the polyp when it is not). In this context, a good polyp segmenta‐ tion method should provide higher Precision and Recall results (Fig. 19 (b)); a method provid‐ ing high Precision with low Recall will provide regions that cannot be used for further polyp characterizationas theycontainlotsofnon-polypinformation(Fig. 19 (c)).Converselyamethod providing with high Recall but low Precision values will be useful for polyp description but

> (a) (b) (c) (d) Figure 19: Interpretation of segmentations: (a) original ground truth. Segmentation results with (a) good Precision and Recall values; (c) good Precision but low Recall value; and (d) low Precision but good Recall value. Mask representing the output of a given method is represented in blue.

**Figure 19.** Interpretation of segmentations: (a) Original ground truth. Segmentation results with (a) good Precision and Recall values; (c) good Precision but low Recall value and (d) low Precision but good Recall value. Mask representing

As for the case of polyp localization, ground truth for polyp segmentation should consists of binary masks representing either the area of the image that is occupied by the polyp -Figure

of the polyp presence likelihood image inside the polyp mask.

the image that is occupied by the polyp, as it is shown in Figure 18.

the image to be used for lesion classification purposes.

will leave a lot of useful polyp content out of posterior analysis (Fig. 19 (d)).

used for lesion classification purposes.

analysis (Fig. 19 (d)).

*Ground truth:*

18 (b)- or the contour of the polyp region -Figure 18 (c)-.

*Ground truth*:

the output of a given method is represented in blue.

*Performance metrics:*

*Ground truth:*

134 Screening for Colorectal Cancer with Colonoscopy

*Ground truth*:

*Performance metrics*:

*GL FL*

(Bernal et al. (2015)). Considering these two metrics, LAcc and concentration, a good localization method should provide a low number of FL while concentrating the majority of the polyp presence likelihood image inside the polyp mask.

Ground truth for polyp localization should consist of binary masks representing

In this case we can have two different types of evaluation, depending on the number of possible classes that we define: if a polyp can only have two different classes we could evaluate our method by checking whether the output of a method coincides or not with the ground truth; in this case for each image we will have a correct (OK) or incorrect classification (NOK). The accuracy of the system will be calculated as

$$Acc = \frac{\text{OK}}{\text{OK} + \text{NOK}}$$

The second type of evaluation is related to multiclass classification; in this case we can also include studies regarding which classes are more easily identified and which classes are mostly confused over each other. In this last case we can use confusion matrices, similar to the ones presented in [54] to represent the output of a given classification method.

#### *Ground truth*:

Ground truth for polyp classification should consist of a label associated to each frame with a polyp; this label must include the given polyp in any of the possible classes defined in the problem.

### **7. Conclusions**

Collaboration between clinicians and computer scientists is crucial for the development of intelligent systems for colonoscopy. Those systems need to be designed to solve real clinical problems if they want to be deployed in clinical environments. Considering this, apart from application development and validation, efforts must be focused on the definition of the aim of the proposed intelligent system.

We have presented in this chapter some of the problems that colonoscopy still present nowadays, being polyp miss-rate the most important of them. Additionally there is a need expressed by clinicians of systems that can allow them to have a first approach to polyp histology, which could be useful to take in-vivo decisions. Considering this we define three possible domains of application of a given intelligent system: polyp detection and localization, polyp classification and development of navigation-assisting and patient follow-up methods.

Once the clinical need is defined, computer scientists must deal with image processing in order to provide with meaningful results. In this context, we have subdivided this problem in two: image preparation for optimal image processing and endoluminal scene description for intelligent system applications.

Regarding image preparation, one of the main objectives of this chapter was to rise up some concerns about image quality for later processing and clinicians and computer scientists must reach an agreement to obtain images that are useful for both domains. Endoluminal scene description has been proven as a challenging task due to the great variability in structures' appearance throughout different interventions. The majority of bibliographical sources are devoted to polyp characterization, although we have observed an increasing interest in the definition of other elements of the scene, as they have been proven to have an impact in polyp characterization tasks. At this point it is important to mention that there are some aspects that we have not covered in full such as patient preparation although it has a direct consequence on the output of a given intelligent system. In this case we opt to follow the same criteria that clinicians do: if patient preparation is bad neither computer vision nor clinicians would be able to distinguish anything.

The objective of the development of an intelligent system is to take profit of the syner‐ gies between clinicians and computer scientists. During the development of a given system, clinicians must provide with data in order to test different methods. We propose in this chapter a validation framework which covers topics such as database and ground truth creation as well as the definition of performance metrics. The proposal of a validation framework including database creation and management along with the definition of standard evaluation metrics can pave the way for a standardized comparison of the performance of intelligent systems which would allow in the future clinicians choose the one that fulfills better their necessities.

The main conclusion that can be extracted from this chapter is that there is indeed room and necessity for the collaboration between these two domains of research. Acknowledg‐ ing the necessities of each other is meant to play a key role in the development of applicable and deployable intelligent systems for colonoscopy.

#### **Author details**

Jorge Bernal1\*, F. Javier Sánchez1 , Cristina Rodríguez de Miguel2 and Gloria Fernández-Esparrach2

\*Address all correspondence to: jbernal@cvc.uab.es

1 Computer Science Department at Universitat Autònoma de Barcelona and Computer Vision Center, Barcelona, Spain

2 Endoscopy Unit, Gastroenterology Department, Hospital Clínic, IDIBAPS, CIBEREHD, University of Barcelona, Barcelona, Spain

### **References**

image preparation for optimal image processing and endoluminal scene description for

Regarding image preparation, one of the main objectives of this chapter was to rise up some concerns about image quality for later processing and clinicians and computer scientists must reach an agreement to obtain images that are useful for both domains. Endoluminal scene description has been proven as a challenging task due to the great variability in structures' appearance throughout different interventions. The majority of bibliographical sources are devoted to polyp characterization, although we have observed an increasing interest in the definition of other elements of the scene, as they have been proven to have an impact in polyp characterization tasks. At this point it is important to mention that there are some aspects that we have not covered in full such as patient preparation although it has a direct consequence on the output of a given intelligent system. In this case we opt to follow the same criteria that clinicians do: if patient preparation is

bad neither computer vision nor clinicians would be able to distinguish anything.

The objective of the development of an intelligent system is to take profit of the syner‐ gies between clinicians and computer scientists. During the development of a given system, clinicians must provide with data in order to test different methods. We propose in this chapter a validation framework which covers topics such as database and ground truth creation as well as the definition of performance metrics. The proposal of a validation framework including database creation and management along with the definition of standard evaluation metrics can pave the way for a standardized comparison of the performance of intelligent systems which would allow in the future clinicians choose the

The main conclusion that can be extracted from this chapter is that there is indeed room and necessity for the collaboration between these two domains of research. Acknowledg‐ ing the necessities of each other is meant to play a key role in the development of applicable

, Cristina Rodríguez de Miguel2

1 Computer Science Department at Universitat Autònoma de Barcelona and Computer

2 Endoscopy Unit, Gastroenterology Department, Hospital Clínic, IDIBAPS, CIBEREHD,

and

intelligent system applications.

136 Screening for Colorectal Cancer with Colonoscopy

one that fulfills better their necessities.

**Author details**

Jorge Bernal1\*, F. Javier Sánchez1

Vision Center, Barcelona, Spain

University of Barcelona, Barcelona, Spain

Gloria Fernández-Esparrach2

and deployable intelligent systems for colonoscopy.

\*Address all correspondence to: jbernal@cvc.uab.es


(FICE). International journal of colorectal disease, 28(11), 1511-1516. DOI:10.1007/ s00384-013-1735-4


a population-based study. Gastroenterology, 146(4), pp. 950-960. DOI:10.1053/ j.gastro.2014.01.013

[22] Kudo, S. E., Wakamura, K., Ikehara, N., Mori, Y., Inoue, H., & Hamatani, S. (2011). Diagnosis of colorectal lesions with a novel endocytoscopic classification–a pilot study. Endoscopy, 43(10), 869. DOI: 10.1055/s-0030-1256663

(FICE). International journal of colorectal disease, 28(11), 1511-1516. DOI:10.1007/

[12] Hoffman, A., Kagel, C., Goetz, M., Tresch, A., Mudter, J., Biesterfeld, S. et al. (2010). Recognition and characterization of small colonic neoplasia with high-definition co‐ lonoscopy using i-Scan is as precise as chromoendoscopy. Digestive and Liver Dis‐

[13] Bernal, J., Śanchez, F. J., Ferńandez-Esparrach, G., Gil, D., Rodŕıguez, C., & Vilariño, F. (2015). WM-DOVA Maps for Accurate Polyp Highlighting in Colonoscopy: Vali‐ dation vs. Saliency Maps from Physicians. Computerized Medical Imaging and

[15] Pellise, M., Fernández-Esparrach, G., Cardenas, A., Sendino, O., Ricart, E., Vaquero et al. (2008). Clinical impact of wide-angle, high-resolution endoscopy in the diagno‐ sis of colorectal neoplasia in a non-selected population: a prospective randomized controlled trial. Gastrointestinal Endoscopy, 67(5), AB101. DOI :10.1016/j.gie.

[16] Rex, D. K., & Helbig, C. C. (2007). High yields of small and flat adenomas with highdefinition colonoscopes using either white light or narrow band imaging. Gastroen‐

[17] Quintero, E., Castells, A., Bujanda, L., Cubiella, J., Salas, D., Lanas, Á. et al. (2012). Colonoscopy versus fecal immunochemical testing in colorectal-cancer screening. New England Journal of Medicine, 366(8), pp. 697-706. DOI: 10.1056/NEJMoa1108895

[18] Barclay, R. L., Vicari, J. J., Doughty, A. S., Johanson, J. F., & Greenlaw, R. L. (2006). Colonoscopic withdrawal times and adenoma detection during screening colonosco‐ py. New England Journal of Medicine, 355(24), pp. 2533-2541. DOI: 10.1056/

[19] van Rijn, J. C., Reitsma, J. B., Stoker, J., Bossuyt, P. M., van Deventer, S. J., & Dekker, E. (2006). Polyp miss rate determined by tandem colonoscopy: a systematic review. The American journal of gastroenterology, 101(2), pp. 343-350. DOI:10.1111/j.

[20] Bretagne, J. F., Manfredi, S., Piette, C., Hamonic, S., Durand, G., & Riou, F. (2010). Yield of high-grade dysplasia based on polyp size detected at colonoscopy: a series of 2295 examinations following a positive fecal occult blood test in a populationbased study. Diseases of the Colon & Rectum, 53(3), pp. 339-345. DOI: 10.1007/DCR.

[21] Samadder, N. J., Curtin, K., Tuohy, T. M., Pappas, L., Boucher, K., Provenzale, D. et al. (2014). Characteristics of missed or interval colorectal cancer and patient survival:

Graphics, 43,pp. 99-111. DOI :10.1016/j.compmedimag.2015.02.007

[14] Modlin, I. M. (2000). A brief history of endoscopy. MultiMed.

terology, 133(1), pp. 42-47. DOI :10.1053/j.gastro.2007.04.029

s00384-013-1735-4

138 Screening for Colorectal Cancer with Colonoscopy

2008.03.119

NEJMoa055498

1572-0241.2006.00390.x

0b013e3181c37f9c

ease, 42(1), 45-50. DOI :10.1016/j.dld.2009.04.005


dominal Imaging. Computational Challenges and Clinical Opportunities (pp. 9-14). Springer Berlin Heidelberg. DOI: 10.1007/978-3-642-25719-3\_2


[44] Bernal, J., Gil, D., Sánchez, C., & Sánchez, F. J. (2014). Discarding Non Informative Regions for Efficient Colonoscopy Image Analysis. In Computer-Assisted and Robot‐ ic Endoscopy (pp. 1-10). Springer International Publishing. DOI: 10.1007/978-3-319-13410-9\_1

dominal Imaging. Computational Challenges and Clinical Opportunities (pp. 9-14).

[33] Tajbakhsh, N., Gurudu, S. R., & Liang, J. (2014). Automatic Polyp Detection Using Global Geometric Constraints and Local Intensity Variation Patterns. In Medical Im‐ age Computing and Computer-Assisted Intervention–MICCAI 2014(pp. 179-187).

[34] Kang, J., & Doraiswami, R. (2003, May). Real-time image processing system for endo‐ scopic applications. In Electrical and Computer Engineering, 2003. IEEE CCECE 2003. Canadian Conference on (Vol. 3, pp. 1469-1472). IEEE. DOI:10.1109/CCECE.

[35] Burrus, C. S., Gopinath, R. A., & Guo, H. (1998). Introduction to wavelets and wave‐

[36] Bernal, J., Sánchez, F. J., & Vilariño, F. (2010). Feature Detectors and Feature Descrip‐

[37] Ameling, S., Wirth, S., Paulus, D., Lacey, G., & Vilariño, F. (2009). Texture-based pol‐ yp detection in colonoscopy. In Bildverarbeitung für die Medizin 2009(pp. 346-350).

[38] Coimbra, M. T., & Cunha, J. S. (2006). MPEG-7 visual descriptors—contributions for automated feature extraction in capsule endoscopy. Circuits and Systems for Video Technology, IEEE Transactions on, 16(5), 628-637. DOI:10.1109/TCSVT.2006.873158

[39] Park, S. Y., Sargent, D., Spofford, I., & Vosburgh, K. G. (2012). A colon video analysis framework for polyp detection. Biomedical Engineering, IEEE Transactions on, 59(5),

[40] Xu, Y. R., & Zhao, J. (2014). Segmentation of haustral folds and polyps on haustral folds in CT colonography using complementary geodesic distance transformation. Journal of Shanghai Jiaotong University (Science), 19, pp. 513-520. DOI: 10.1007/

[41] Van Wijk, C., Van Ravesteijn, V. F., Vos, F. M., & Van Vliet, L. J. (2010). Detection and segmentation of colonic polyps on implicit isosurfaces by second principal curvature flow. Medical Imaging, IEEE Transactions on, 29(3), pp. 688-698. DOI: 10.1109/TMI.

[42] Bernal, J., Núnez, J. M., Sánchez, F. J., & Vilariño, F. (2014). Polyp Segmentation Method in Colonoscopy Videos by Means of MSA-DOVA Energy Maps Calculation. In Clinical Image-Based Procedures. Translational Research in Medical Imaging (pp.

[43] Ganz, M., Yang, X., & Slabaugh, G. (2012). Automatic segmentation of polyps in colo‐ noscopic narrow-band imaging data. Biomedical Engineering, IEEE Transactions on,

41-49). Springer International Publishing. DOI: 10.1007/978-3-319-13909-8\_6

59(8), pp. 2144-2151. DOI: 10.1109/TBME.2012.2195314

tors: Where We Are Now. Technical Report. Computer Vision Center.

Springer Berlin Heidelberg. DOI: 10.1007/978-3-540-93860-6\_70

Springer Berlin Heidelberg. DOI: 10.1007/978-3-642-25719-3\_2

let transforms (Vol. 998). New Jersey: Prentice hall.

pp. 1408-1418. DOI: 10.1109/TBME.2012.2188397

2003.1226181

140 Screening for Colorectal Cancer with Colonoscopy

s12204-014-1534-2

2009.2031323

Springer International Publishing. DOI: 10.1007/978-3-319-10470-6\_23


## *Edited by Rajunor Ettarh*

Colorectal cancer remains a major health issue for many developed regions around the world. The good news is that early detection has significantly improved overall survival rates and continues to do so. A number of prevention strategies contribute to this positive trend, and today a patient who undergoes a colonoscopy for screening purposes stands a much better chance of being effectively surveyed for prevention of colorectal cancer. Patients can rely increasingly on the improved datasets and technical advances that are being made in screening approaches and skills. With continued progress, particularly in the partnership between clinicians and computer scientists, the future for colorectal cancer surveillance looks increasingly positive for the development of improved tools and methods.

Photo by carloscastilla / DollarPhoto

Screening for Colorectal Cancer with Colonoscopy

Screening for Colorectal

Cancer with Colonoscopy

*Edited by Rajunor Ettarh*