**2. Basic principles of CAD**

Computer-aided diagnosis or computer-aided detection refers to the combination of computer technology with medical image processing technology and other possible physiological and biochemical techniques to assist doctors for clinical decision-making and improve the accuracy of diagnosis. The computer-aided detection focus on the localization task, while the computer-aided diagnosis focuses on characterization task, such as the distinction between benign and malignant tumors, and classification among different tumor types. The computer only needs to annotate the abnormal signs and then carries out conventional image processing

**3**

**Figure 2.**

*Flowchart of computer-aided diagnosis [11].*

*Introductory Chapter: Computer-Aided Diagnosis for Biomedical Applications*

without further diagnosis. In other words, computer-aided diagnosis is the extension and ultimate goal of computer-aided detection. Correspondingly, computeraided detection is the basis and necessary stage of computer-aided diagnosis. Previous studies proved that CAD plays a significant role in improving the sensitiv-

**Figure 1** shows the flowchart of the use of computer-aided detection in medical images for the diagnosis of lesions. The model consists of four main steps and two optional steps [11]. This first step makes the rest of the steps focus on the organ. The second step obtains a high sensitivity level because the sensitivity lost in this step cannot be recovered in the later steps. The third step aims to segmentation and feature analysis of the detected lesions. The fourth step aims to classify the segmented lesions by extracting pattern features such as gray-level-based features, texture features, and morphologic features. A machine-learning technique has been applied in this step. The machine-learning technique determines "optimal" boundaries for separating classes in the multidimensional feature space [12]. The final step classifies the detected lesions into the lesion and non-lesion groups. This step determines the final performance of the computer-aided detection scheme. The optional step 1 is applied to improve the performance of the system after the first step, which helps to improve the sensitivity and specificity of the detection in the subsequent step. The optional step 2 is usually applied at the end of the system to reduce false-positive rates, which could improve the specificity of the proposed

**Figure 2** shows the use of computer-aided diagnosis in medical images for diagnosis of lesions. The first step aims to detect lesions automatically or manually. The second step aims to segmentation and feature analysis of the detected lesions. The detected lesions are first segmented, and pattern features are extracted from the segmented lesions. With the extracted features, the lesions are classified into benign or malignant or different lesion types. This system may determine the likelihood of

*DOI: http://dx.doi.org/10.5772/intechopen.88835*

ity, specificity, and accuracy of diagnosis.

being malignancy or a specific type of lesion.

CAD system.

**Figure 1.**

*Flowchart of computer-aided detection [11].*

#### *Introductory Chapter: Computer-Aided Diagnosis for Biomedical Applications DOI: http://dx.doi.org/10.5772/intechopen.88835*

without further diagnosis. In other words, computer-aided diagnosis is the extension and ultimate goal of computer-aided detection. Correspondingly, computeraided detection is the basis and necessary stage of computer-aided diagnosis. Previous studies proved that CAD plays a significant role in improving the sensitivity, specificity, and accuracy of diagnosis.

**Figure 1** shows the flowchart of the use of computer-aided detection in medical images for the diagnosis of lesions. The model consists of four main steps and two optional steps [11]. This first step makes the rest of the steps focus on the organ. The second step obtains a high sensitivity level because the sensitivity lost in this step cannot be recovered in the later steps. The third step aims to segmentation and feature analysis of the detected lesions. The fourth step aims to classify the segmented lesions by extracting pattern features such as gray-level-based features, texture features, and morphologic features. A machine-learning technique has been applied in this step. The machine-learning technique determines "optimal" boundaries for separating classes in the multidimensional feature space [12]. The final step classifies the detected lesions into the lesion and non-lesion groups. This step determines the final performance of the computer-aided detection scheme. The optional step 1 is applied to improve the performance of the system after the first step, which helps to improve the sensitivity and specificity of the detection in the subsequent step. The optional step 2 is usually applied at the end of the system to reduce false-positive rates, which could improve the specificity of the proposed CAD system.

**Figure 2** shows the use of computer-aided diagnosis in medical images for diagnosis of lesions. The first step aims to detect lesions automatically or manually. The second step aims to segmentation and feature analysis of the detected lesions. The detected lesions are first segmented, and pattern features are extracted from the segmented lesions. With the extracted features, the lesions are classified into benign or malignant or different lesion types. This system may determine the likelihood of being malignancy or a specific type of lesion.

#### **Figure 2.**

*Flowchart of computer-aided diagnosis [11].*

*Computer Architecture in Industrial, Biomechanical and Biomedical Engineering*

dual dilemma existed until the late 1980s and early 1990s. Thanks to the rapid development of computer, mathematics, and statistics technologies, CAD has been improved qualitatively. In recent years, the CAD system has been rapidly applied in medical imaging and has made gratifying achievements in some developed

Computer-aided diagnosis or computer-aided detection refers to the combination of computer technology with medical image processing technology and other possible physiological and biochemical techniques to assist doctors for clinical decision-making and improve the accuracy of diagnosis. The computer-aided detection focus on the localization task, while the computer-aided diagnosis focuses on characterization task, such as the distinction between benign and malignant tumors, and classification among different tumor types. The computer only needs to annotate the abnormal signs and then carries out conventional image processing

**2**

**Figure 1.**

*Flowchart of computer-aided detection [11].*

countries.

**2. Basic principles of CAD**
