Characterization of Brain Stroke Using Image and Signal Processing Techniques

*Abdullah Alamoudi and Yousif Abdallah*

## **Abstract**

Cross-sectional imaging approaches play a key role in assessing bleeding brain injuries. Doctors commonly determine bleeding size and severity in CT and MRI. Separating and identifying artifacts is extremely important in processing medical images. Image and signal processing are used to classify tissues within images closely linked to edges. In CT images, a subjective process takes a stroke 's manual contour with less precision. This chapter presents the application of both image and signal processing techniques in the characterization of Brain Stroke field. This chapter also summarizes how to characterize the brain stroke using different image processing algorithms such as ROI based segmentation and watershed methods.

**Keywords:** brain stroke, image processing, signal processing

## **1. Introduction**

A medical condition called a stroke is cell death associated with weak blood flow into the brain. It is an ischemic stroke caused by insanity caused by blood flow, and it is a bleeding that in two different forms causes hemorrhagic stroke. Both have helped to prevent the proper functioning of brain sections [1–7]. The World Health Organization has indicated that it is known as a neurological impairment by cerebral vascular factor as a stroke in the 1970s. One of the most common types of stroke is ischemia which causes a loss of brain functions because of severe lack of blood supply [8–11]. Because there are treatments that can reduce the severity of stroke symptoms if given early enough, many people now prefer the more urgent wording "stroke" or "acute ischemic stroke (AIS)" over the less urgent "brain attack" and "acute ischemic stroke" [12–19]. The headaches caused by this disease can be severe and they may cause strokes. In a stroke, symptoms might include possible partially done strokes. Problems that can arise, if an e-cigarette is not used properly, include bladder sickness and bladder dysfunction. High blood pressure of the veins is the most important risk factor on stroke. The risk of the disease falls on other risk factors, such as smoking, obesity, elevated blood cholesterol, diabetes, and hypertension [20–24]. Diagnosis is usually based on the clinical investigation and is assisted by medical imaging, such as CT or MRI scans [25–29]. CT scans may rule out bleeding, but ischemia, which usually does not appear early in the CT scan, may not necessarily be excluded [30–33]. Other tests are performed to identify risk factors and to rule out other possible causes, such as electrocardiogram (ECG)

and blood tests [34–37]. Aspirin may be used to prevent clots from forming on the inside of the blood vessel walls. Statins may be used to prevent clots from forming within your blood vessels. For small areas of injury, clot removal may be performed. Stroke reminiscence may also be performed to create a place where the brain can begin again with stroke recovery. If your brain is injured, a surgically placed clot of another substance may be used to stop bleeding in the parts of the brain that are injured [7, 38]. Personal computer with suitable software can use to modify and custom a digital image algorithm. A personal computer can produce your own customized digital image. Digital image processing, as opposed to analog signal processing, has many advantages. The new software has characteristics that allow one to be more flexible with visually, audio and text output features. Because digital image processing can be modeled in a multidimensional way, digital images can be defined as two (or more) dimensional. In 1972, EMI Hounsfield, (a gentleman engineer) calculated subtle methods to immerse a radiolucent X-ray tube into a potable medium which was used to take images within patients' heads (computer tomography). In order to construct a cross-section of the entirety of an image, the CT scanner is established on the basis of the anatomy of the human body. In 1975, EMI demonstrated a full body scanner with a cross-sectional scan in order to identify body parts and other people in the body. The Nobel Prize is earned by this type of research in 1979. In the 1990s, medical image processing technologies were used to control heart rates during coronary artery bypass grafts [17, 24].

## **2. Stroke classification**

These two types of strokes, Ischemic and Hemorrhagic, have different properties. They are caused by a stroke, and the damage that stroke brings to brain. Besides, the infections that cause meningitis are caused when blood vessels burst or leaks out of the body. 87% of stroke pain is due to a reduced blood supply to the brain, and the rest is due to bleeding. Some people living in ischemic areas might feel some internal bleeding. This is a cause of poor blood flow in the walls of the heart [2, 39].

Strokes that are the result of ischemic damage to the brain can be caused. In order to answer this question, one will need to weigh four different points of view:


Acute Ischemic Strokes (IS) has a myriad of classifications. The most common one is Cryptogenic stroke. it comprises the usual cause for this type of stroke which makes up about 30–40 percent of all ischemic strokes. The number of IS strokes is about 1,500,000 per year. Depending on the level of symptoms, the Oxford Community stroke program (SCAP) consists of symptomatic or refractory (traumatic) strokes; these experiences are considered to be kind of those seen in severe stroke; the Mini–Mental state exam, or the McIntosh Edinburgh assessment; the

*Characterization of Brain Stroke Using Image and Signal Processing Techniques DOI: http://dx.doi.org/10.5772/intechopen.96288*

total anterior circulation, or lacuna infarction; or secondary circulatory event is an integral part of the Oxford Community stroke program, therefore, the focal branch infarction, or focal ischemic attack or stroke; and the changes in mental, depending on the amount of signs and symptoms of the Oxford Community stroke program (SCAP), and the Oxford Community stroke program (SCAP); those mentioned are the "Oxford Community stroke program", the "Oxford Community stroke program" attack! (POCI). The level of the stroke, the area of the brain affected, the symptoms and prognosis are defined by the following four groups; [24, 25] The TOAST classification (Orga Stroke Assay 10172 Acute Stroke Therapy) is dependent on clinical symptoms and the outcomes of further investigations [41–43]. Good thing is that a stroke is known to be partly caused by particular risk factors, such as thrombosis or embolism of arteriosclerosis or stroke [27, 44–47].

There are two types of strokes that cause bleeding: hemorrhagic and ischemic.


There are different types of bleeding, one of which is ablated. There are also two primary kinds of intracranial hemorrhages, which are also called "bleeding" or "bleeding." Included under that are two forms of "bleeding" called "hematoma," which is a type of subdural hematoma. The "whole" of this document is worded confusingly. It says "'hematoma" is also included under" 'bleeding." It should read "Also, there are two different kinds of hemorrhagic strokes…" [48–51].

## **2.1 Stroke signs and symptoms**

On the contrary, stroke symptoms often arrive without prior warning. When you see these symptoms, it's probably not going to be long before they become noticeable. In some cases, symptoms vary in different brain areas because of an infection. This disease has more effect in larger parts of the brain, meaning that the loss of neural features will likely be more costly. In addition, there are different signs that may lead to strokes. This part of the patient may compress in the location of the bleeding, such as in intracranial bleeding. For some people who have certain types of stroke, headaches are not always associated with their stroke. For example, subarachnoid hemorrhage and cerebral thrombosis are strokes that sometimes cause headaches [36, 52, 53].

If it uses the main central pathways that the brain is involved with, it will work. When we look at a specific test, we can find that its one of the three pathways we use can fail.


## *Biomedical Signal and Image Processing*

Most of the time, symptoms on one side of the body are involved (unilateral). Since the damage to the brain is always located in another hemisphere, it affects the non-damaged side of the brain instead of the damage side. Although some signs may be caused by bumps in the legs, such as leg pain, it is absolutely clear that any serious injury can affect all of these symptoms. Therefore, to approach whether or not there is also a symptom that is not definite is difficult. To man into. Therefore, the part of the brain that is damaged by the stroke can cause deficits in the following cranial nerve functions:


## *2.1.1 Stroke symptoms*


If the cerebellum is busy and working well, a person will have ataxia.

*Characterization of Brain Stroke Using Image and Signal Processing Techniques DOI: http://dx.doi.org/10.5772/intechopen.96288*


## **2.2 Stroke causes**

## *2.2.1 Thrombotic stroke*

When a child has a thrombosis, there is typically a plaque deposit around the blockage. Stroke may not be as serious as a heart attack, since the blood pressure may not necessarily drop like in a heart attack. It is an embolism in the heart (see below). This medication, applied to a vein, could lead to the development of a blood clot, as this may make the blood flow less freely. Both types of stroke can result in a pronounced stroke.


## *2.2.2 Embolic stroke*

The reason why a stroke occurs is because a clot gets in the blood vessels. The doctor will give you some medication (a blockage of an artery). There were many blood clots found in the different vessels, with these being composed of the most of them being blood fat cells and bacterial accumulations. It is always important to be able to locate and stop the clot. Patients who have this disease may start to experience symptoms within just a couple of weeks after the symptoms appear. These symptoms can be just as intermittent as the embolus re-routes elsewhere or goes back empty. The nail used in this study may also be applicable to patients with coronary artery disease (CAD) and atrial fibrillation (AF) [44].

With stroke, the heart, brain, and blood vessels each need to be evaluated separately [65].


If a stroke occurs on the same side of the body, it predicts that the risk of a stroke will be 1% a year. The cause of stroke in this kind of patients is past a particular form of the disease (ESUS). The subtypes of cryptogenic stroke. The first part is likely that one of the ischemic strokes will be [65–67].

## *2.2.3 Cerebral hypoperfusion*

When the brain is not given enough blood supply, it is called mild corto-cortical hypoperfusion. Sometimes, part of the brain is lost to various circumstances, such as a someone being concussed. Heart diseases, such as a recent heart attack, is a common health concern. This is the reason. For the purpose of. According to the studies, all areas of the brain has grown and is now much larger. However, at some point it might seem small, like compressed. Development and the brain It is a form of heart attack called water fall. When the brain is under a lot of strain, blood can cease from flowing at the pained area, and brain damage can be caused. When the brain is in a stress condition, brain activity will be reduced or stifled. The cerebral venous sinus is not only dangerous for forming clots in the brain, but also dangerous for how it raises blood pressure. This type of stroke is much more likely than any other type of stroke [22]. Strokes can occur to the brain or to the arteries because of a damage that can take place in either location. There are a number of potential causes to understand in this patient setup. The possibility of a medical condition present is the most likely to be possible cause (e.g., amphetamines or cocaine). The blood cells migrate to the ventricle in an unhealthy state. Preventing this can prevent tumor cells from becoming engorged. About a third of the ventricles are not being used by the brain. Within 30 days of an intracerebral hemorrhage, at least a 40% survival rate is possible and higher than that of other neurodegenerative diseases and subarachnoid hemorrhages (which technically may also be classified as a type of stroke [68, 69]). Heart attacks can also result from other conditions that occur, due to the heart attack itself. Most likely, cocaine is the reason behind an increase in crack use [49].

### *2.2.4 Silent stroke*

Apart from damage, this disease also has several silent strokes that can cause longterm brain damage. Tests have shown that silent strokes can sometimes lead to a disease in the future. All people who have experienced a major stroke are also susceptible to having a silent, milder stroke. The lifetime risk of having a stroke is slightly higher than the odds of getting killed during a car accident. Of the 770,000 strokes, about half of them were symptomatic strokes such as blood clots, and 11 million were strokes, hemorrhages, or MRI infarctions. Injury detection by such a system, which is so advanced, often is not noticed. Those strokes that are diagnosed as silent occur five times more often than those strokes that are fatal. When the body is working out, it is automatically producing the malodorous by-product of metabolism, the infamous lactic acid. Acidosis occurs when too much acid is produced, which can interfere with the normal functioning of the mind. Since this is an area in which the industry demonstrated great innovation, with fat in the form of ethersys which continued to cause companies to pull billions of dollars of income from emerging markets their findings were thereafter patented. They made patents filings which include the discovery of a variety of new compounds related to adenosine which leads them to prevent competitors from successfully copying the source of revenue (ATP). The part of town is referred to as simply "Mid" because it is inconveniently located. Glutamate binds to receptors to tear down the neurons. Neural ion homeostasis is usually maintained by a few rare plants. This stroke reduces the value, making it harder to chip off the paint, damaging the microscopic paint. In that case, high levels of glutamate are sequestered in the cytoplasm.

### *Characterization of Brain Stroke Using Image and Signal Processing Techniques DOI: http://dx.doi.org/10.5772/intechopen.96288*

The elevated levels of chemical glutamate cause calcium to enter cells. A series of chain reactions in mitochondria and lack of sufficient oxygen over time leads to a steady accumulation of free radicals and reactive oxygen species. When we have finished digesting and removing cellular and extracellular material. There are several potential problems with the interior of the ship. Antioxidants are only effective in the endothelium and do NOT work in the brain itself. Free radicals are involved in cell death, but so is life. The brain is particularly sensitive to lack of blood supply [70–74].

The plasma membrane of damaged cells can lose its blood supply and be unable to uptake oxygen and much needed nutrients. In addition, inflammation can cause damage the overlying brain. It is capable of other possible proteolytic actions. It is possible for severe brain or head injuries to occurs due to a loss of cerebral vascular integrity, or to a case of dangerous stroke. Diseases that cause hemorrhagic strokes have causes that are still being investigated. There are many possible causes of a stroke, such as high blood pressure, aneurysm, a vascular fistula, arterial disease of the carotid artery, and an allergic reaction. When using the e-cigarette, people can have minor damage in the brain due to a lack of blood flow to the brain [61, 74–78].

## **2.3 Stroke diagnosis**

MRI and CT scans are used to determine the amount of brain damage caused when a stroke occurs. The imaging methods allowed the phenomenon of occultation to be located. Scanning techniques can help identify a stroke's subtype and identify where its source is. Doctors may also be able to perform other blood tests that can help diagnose a stroke as well. Blood tests can reveal what something is really like to another person [62].

## *2.3.1 Physical examination*

A medical examination involves cross-referencing a person's past medical history to their current clinical history. If there is a normal level of stroke severity, it is recommended to assign a normal level.

## *2.3.2 Imaging*

The case study: "Ischemic stroke" [63].

	- Recovery Time: (less than 10 percent within first 3 hours of symptom onset)
	- Specificity of positive predictive value = 96%.
	- Affectivity is, well, very strong.
	- Specificity of objects (98%).
	- It's a type of MRI scan for NMA (without contrast enhancements)
	- The sensitivity dropped by eleven percent.

The MRI scan is sufficient to be evaluated for a chronic stroke. However, it may also be necessary for scanning with SPECT or PET. The metabolism of PET, brain blood movement isotopes, PET, and PET neurons are reported as PET and SPECT. In the blood supply, CT scans identify mild ischemic strokes with tender detection of blight. To estimate the underlying cause of a CT test, it is possible to use many other image enhancers. Determination of the quality of these medicines is very important, depending on the prescription. The selection of the stroke response will depend heavily on the cause of the stroke. The conventional way is the following:


If hemorrhagic stroke test deficiencies are present, the CT or MRI contrast scan can identify cerebral artery deformation or any other source of bleeding or structural MRI if this does not result (for example, aneurysms). Invasive cerebral angiography may also be performed if the underlying cause of bleeding is not identified but implies access to the blood stream with the intravascular catheter, which can cause additional strokes and complications at the place of insertion, and therefore this study is reserved for certain situations [65].

### *2.3.3 Misdiagnosis*

The diagnosis ranges from 2 to 27 per cent of ischemic stroke cases. A stroke that has other characters is the 'Chameleon stroke.' Failure to diagnose people without a stroke also has the potential. One to two percent of the total is typically one to two percent of the penile bleeder that produces intracerebral bleeding through thrombosis addiction (one to two percent, less in stroke). The cost of medical treatment is helped by this superfluous treatment. However, the AHA/ASA advice suggests that further studies should delay intravenous tPA treatment [67]. African Americans, Latin Americans, Asians, and Pacific Islanders are more likely to be misrepresented for a condition other than stroke when they have a stroke. Moreover, women younger than 44 are 7 times more likely than people older than 75 years of age to have a stroke. For younger trauma sufferers, this can be the case [66]. Experimental studies in other medical centers have shown that hyperacute RMI m is hyperacute [66].

## **3. Medical imaging processing for stroke**

Doctors have multiple strategies and methods for identifying medical images to diagnose diseases. Cell segmentation in diseases, with the help of MATLAB, can be the easiest way for doctors to examine illnesses of the immoral tissue. Image classification in sectors with structural units is the interpretation and analysis of an image that you plan to be mutually compatible. The MI brain imaging is one of the most competitive physiques in the organism, so explaining brain segmentation is a detailed assignment of which the assistant will not work because… Biomedical imagery represents the body that consists of the soft tissue, organ, and bone structure. In such circumstances, they cannot tell whether or not this raw biomedical view of adolescent health climaxes. This means that biomedical image analysis through image handling is so critical that it is essential for medics to be conscious of special problems and of their development related to disease and growth problems. There are numerous types of medical visual sectors in MATLAB application like threshold and morphology [5, 12]. Morphological and philological methods are forms of different pictures for some brain strokes. The simplest and the most frequently identified technique in the pixel is to decide that for all pixels measured below a given point above or below intensity it would form part of a pixel. Form processing requires a morphological work. This combines two different morphological procedures, expansion, and exhaustion. With a dilate, center activation, or for the active expansion of pixel, with erosion of their neighbors is the minimum and, at a minimum, the minimum required to accommodate its neighbors. The two activities are linear as well. They are not settings of rotatable transitions which usually result in an unusual image from each of them, one after another. It's used to prevent background defects from deepension to dilation (Opening). However, by closing small gaps [5, 6], to compensate the shape of the object, erosion dilation will also be used to fill small gaps. It becomes the U.S.'s fourth major cause of death in the United States that kills people every four min. More then00,000 people who experienced a stroke arise from respiratory complications in the brain (a bloodstream blocks) or from the spotted blood vessel. The level of brain improvement, for example, increased owing to an increase in medical malpractice. This has in many ways culminated in an appropriate medical diagnosis. However, it is not simple to correctly detect an incident. In this document we seek to articulate a conceptual process to help physicians deal with illnesses on a social basis. Skin rash was detected through image purchase selectivity. In order to evaluate brain stroke images from magnetic resonance imaging, the examiner uses the MATLAB panel (MRI).

Many scientists such as [11, 12, 35, 56, 71–74] described in this paper, Diagnostic Application of Brain Image Processing and the System of Stroke Analysis, that, particularly relative to standard MRI technologies are generally used for early examination and therapy of vaccines. This study aims at using the photographic editing that can effectively measure volume strength and provide 3D evidence of damage restructuring with a sound analysis system, assisted by software technicians. Add on that. Our system envisages increased accuracy in the design and interpretation of acute stroke damage, longer calculation time and volume, and a 3D repair plan that will make doctors simpler for dealing with and better evaluating patients, as well as to treat them better. The study findings have shown that this suggested system is better efficient than the existing technique, so this system is therefore also an efficiency. The case study also illustrated through a quantified ischemic stroke study in their picture analysis that the ischemic stroke keeps it producing. This will lead to the murder of blood bacteria on the cells of the brain. Imaging of magnetic resonance scans was widely used to identify brain magnetic strokes. An independent authority is in this study which may use the image processing techniques to

diagnose stroke. There are six steps in this part. In the first step, MRI images retain knowledge. Pre-processing and filtrating should be carried out on raw information. In the third time period, the acquisition of a symmetrical image is traced by Midline. The image is bifurcated in the third step. In addition, neural networks are assigned to Phase 2 during the fifth phase of Imaging quality patterns in physical texture measurements, while normal and infected cerebral physiques have been set aside for the sixth. The downside is that in the early stages they can trace strokes. The approach of a project is simple and time consuming, as well as efficient [79].

## **4. Conclusions**

In the chapter, a process was demonstrated for segmenting medical images of a stroke. Brain stroke segmentation, including pixels and morphology, consists of several segmentation steps and is a dynamic phase. The best way of exploring the MATLAB stroke with MRI images is to use clear images with more intellectual information provided by the MRI unit. Such approaches can discover different types of strokes at different levels of the brain. Researchers hope that, through this project, physicians will be able to understand MRI stroke photos.

## **Acknowledgements**

This research was reinforced by the Applied Medical Science College of Majmaah University, Majmaah 11952, P.O box 1405, Saudi Arabia. We would like to express our gratitude toward RA (r.albaradie@mu.edu.sa) supervisor of the Stroke chair, Majmaah University and Dr. Khalid Bin Saad Al Murqrin, Rector, Majmaah University for providing the necessary support and assistance for completing this study This work was carried out under project number 38/130 Majmaah university.

## **Conflict of interest**

The authors declare no conflict of interest.

## **Author details**

Abdullah Alamoudi and Yousif Abdallah\* Department of Radiological Science and Medical Imaging, College of Applied Medical Science, Majmaah University, Saudi Arabia

\*Address all correspondence to: y.yousif@mu.edu.sa

© 2021 The Author(s). Licensee IntechOpen. 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.

*Characterization of Brain Stroke Using Image and Signal Processing Techniques DOI: http://dx.doi.org/10.5772/intechopen.96288*

## **References**

[1] G. Agam, D. Weiss, M. Soman, K. Arfanakis Probabilistic Brain Lesion Segmentation in DT-MRI (2007), pp. 89-92

[2] Ames III, R.L. Wright, M. Kowada, J.M. Thurston, G. Majno Cerebral ischemia. II. The no-reflow phenomenon The American Journal of Pathology, 52 (1968), pp. 437-453

[3] Anon.New resolution strategy for multi-scale reaction waves using time operator splitting and space adaptive multiresolution: application to human ischemic stroke, volume 34, EDP Sciences (2011)

[4] J.F. Arenillas, A. Rovira, C.A. Molina, E. Grivé, J. Montaner, J. Alvarez-SabínPrediction of early neurological deterioration using diffusion- and perfusion-weighted imaging in hyperacute middle cerebral artery ischemic stroke, Stroke, 33 (2002), pp. 2197-2203

[5] J. Astrup, L. Symon, N.M. Branston, N.A. LassenCortical evoked potential and extracellular K + and H + at critical levels of brain ischemia, Stroke, 8 (1977), pp. 51-57

[6] J. Astrup, B.K. Siesjö, L. SymonThresholds in cerebral ischemia — the ischemic penumbra

[7] Stroke, 12 (1981), pp. 723-725

[8] H. Bagher-Ebadian, K. Jafari-Khouzani, P.D. Mitsias, M. Lu, H. Soltanian-Zadeh, M. Chopp, J.R. EwingPredicting final extent of ischemic infarction using artificial neural network analysis of multiparametric MRI in patients with stroke, PloS One, 6 (2011), p. e22626

[9] O.Y. Bang, J.L. Saver, B.H. Buck, J.R. Alger, S. Starkman, B. Ovbiagele, D. Kim, R. Jahan, G.R. Duckwiler, S.R.

Yoon, F. Viñuela, D.S. Liebeskind, UCLA Collateral InvestigatorsImpact of collateral flow on tissue fate in acute ischaemic stroke. Journal of Neurology, Neurosurgery, and Psychiatry, 79 (2008), pp. 625-629

[10] P.A. Barber, D.G. Darby, P.M. Desmond, Q. Yang, R.P. Gerraty, D. Jolley, G.A. Donnan, B.M. Tress, S.M. DavisPrediction of stroke outcome with echoplanar perfusion- and diffusionweighted MRI

[11] Neurology, 51 (1998), pp. 418-426

[12] R. Bonita Epidemiology of stroke Lancet, 339 (1992), pp. 342-344

[13] J. Braun, J. Bernarding, H.-C. Koennecke, K.-J. Wolf, T. TolxdorffFeature-based, automated segmentation of cerebral infarct patterns using T2- and diffusionweighted imaging. Computer Methods in Biomechanics and Biomedical Engineering, 5 (2002), pp. 411-420

[14] T. Brott, J.R. Marler, C.P. Olinger, H.P. Adams Jr., T. Tomsick, W.G. Barsan, J. Biller, R. Eberle, V. Hertzberg, M. WalkerMeasurements of acute cerebral infarction: lesion size by computed tomography. Stroke, 20 (1989), pp. 871-875

[15] J.H. Burdette, A.D. Elster, P.E. RicciAcute cerebral infarction: quantification of spin-density and T2 shine-through phenomena on diffusionweighted MR images. Radiology, 212 (1999), pp. 333-339

[16] R.A. Carano, K. Takano, K.G. Helmer, T. Tatlisumak, K. Irie, J.D. Petruccelli, M. Fisher, C.H. SotakDetermination of focal ischemic lesion volume in the rat brain using multispectral analysis. Journal of Magnetic Resonance Imaging, 8 (1998), pp. 1266-1278

[17] G. Chapuisat, M.A. Dronne, E. Grenier, M. Hommel, H. Gilquin, J.P. BoisselA global phenomenological model of ischemic stroke with stress on spreading depressions. Progress in Biophysics and Molecular Biology, 97 (2008), pp. 4-27

[18] G. Chapuisat, M.-A. Dronne, E. Grenier, M. Hommel, J.-P. BoisselIn silico study of the influence of intensity and duration of blood flow reduction on cell death through necrosis or apoptosis during acute ischemic stroke. Acta Biotheoretica, 58 (2010), pp. 171-190

[19] M. Chawla, S. Sharma, J. Sivaswamy, L. KishoreA method for automatic detection and classification of stroke from brain CT images Engineering in Medicine and Biology Society, 2009. EMBC 2009. Annual International Conference of the IEEE. IEEE (2009), pp. 3581-3584

[20] T. Chemmanam, B.C.V. Campbell, S. Christensen, Y. Nagakane, P.M. Desmond, C.F. Bladin, M.W. Parsons, C.R. Levi, P.A. Barber, G.A. Donnan, S.M. Davis, EPITHET InvestigatorsIschemic diffusion lesion reversal is uncommon and rarely alters perfusion–diffusion mismatch. Neurology, 75 (2010), pp. 1040-1047

[21] P. Choi, V. Srikanth, T. Phan"Fogging" resulting in normal MRI 3 weeks after ischaemic stroke. BMJ Case Reports, 2011 (2011)

[22] L. Contin, C. Beer, M. Bynevelt, H. Wittsack, G. GarridoSemiautomatic segmentation of core and penumbra regions in acute ischemic stroke: preliminary results. IWSSIP International Conference (2010)

[23] S.B. Coutts, J.E. Simon, A.I. Tomanek, P.A. Barber, J. Chan, M.E. Hudon, J.R. Mitchell, R. Frayne, M. Eliasziw, A.M. Buchan, A.M. DemchukReliability of assessing percentage of diffusion–perfusion

mismatch. Stroke, 34 (2003), pp. 1681-1683

[24] K.A. Dani, R.G.R. Thomas, F.M. Chappell, K. Shuler, K.W. Muir, J.M. WardlawSystematic review of perfusion imaging with computed tomography and magnetic resonance in acute ischemic stroke: heterogeneity of acquisition and postprocessing parameters: a translational medicine research collaboration multicentre acute stroke imaging study. Stroke, 43 (2012), pp. 563-566

[25] P. Dastidar, T. Heinonen, J.P. Ahonen, M. Jehkonen, G. MolnárVolumetric measurements of right cerebral hemisphere infarction: use of a semiautomatic MRI segmentation technique. Computers in Biology and Medicine, 30 (2000), pp. 41-54

[26] B. Dawant, A. ZijdenbosImage segmentation. Handbook of Medical Imaging, 2 (2000), pp. 71-127

[27] M.-A. Dronne, J.-P. Boissel, E. Grenier, H. Gilquin, M. Cucherat, M. Hommel, E. Barbier, G. BriccaMathematical modelling of an ischemic stroke: an integrative approach. Acta Biotheoretica, 52 (2004), pp. 255-272

[28] M.-A. Dronne, J.-P. Boissel, E. GrenierA mathematical model of ion movements in grey matter during a stroke

[29] Journal of Theoretical Biology, 240 (2006), pp. 599-615

[30] T. Dumont, M. Duarte, S. Descombes, M. Dronne, M. Massot, V. LouvetSimulation of human ischemic stroke in realistic 3D geometry: a numerical strategy. Submitted to Bulletin of Math, Biology (2010)

[31] J. Duncan, N. AyacheMedical image analysis: progress over two decades

*Characterization of Brain Stroke Using Image and Signal Processing Techniques DOI: http://dx.doi.org/10.5772/intechopen.96288*

and the challenges ahead, pattern analysis and machine intelligence, IEEE Transactions on, 22 (2000), pp. 85-106

[32] V. Duval, S. Chabaud, P. Girard, M. Cucherat, M. Hommel, J.P. BoisselPhysiologically based model of acute ischemic stroke Journal of Cerebral Blood Flow and Metabolism, 22 (2002), pp. 1010-1018

[33] M.G. Dwyer, N. Bergsland, E. Saluste, J. Sharma, Z. Jaisani, J. Durfee, N. Abdelrahman, A. Minagar, R. Hoque, F.E. Munschauer III, R. ZivadinovApplication of hidden Markov random field approach for quantification of perfusion/diffusion mismatch in acute ischemic stroke. Neurological Research, 30 (2008), pp. 827-834

[34] G. Erus, E. Zacharaki, N. Bryan, C. DavatzikosLearning high-dimensional image statistics for abnormality detection on medical images. Computer Vision and Pattern Recognition Workshops (CVPRW), 2010 IEEE Computer Society Conference on. IEEE (2010), pp. 139-145

[35] Ford, H. An, K.V. LinDefining the ischemic penumbra using hyperacute neuroimaging: deriving quantitative ischemic thresholds. Translational Stroke Research (2012), pp. 1-7

[36] N. Ghosh, R. Recker, A. Shah, B. Bhanu, S.A. ObenausAutomated ischemic lesion detection in a neonatal model of hypoxic ischemic injury. Journal of Magnetic Resonance Imaging, 33 (2011), pp. 772-781

[37] N. Ghosh, Y. Sun, C.T. BhanuComputational analysis: a bridge to translational stroke treatment. Translational Stroke Research (2012), pp. 881-909

[38] V. Gupta, B. Prakash, W.L. NowinskiAutomatic and rapid identification of infarct slices and hemisphere in DWI scans. Academic Radiology, 15 (2008), pp. 24-39

[39] N. Hevia-Montiel, J.R. Jiménez-Alaniz, V. Medina-Bañuelos, O. Yáñez-Suárez, C. Rosso, Y. Samson, S. Baillet Robust nonparametric segmentation of infarct lesion from diffusion-weighted MR images. Conference Proceedings IEEE Engineering in Medicine and Biology Society, 2007 (2007), pp. 2102-2105

[40] E. Grenier, D. Bresch, M. Dronne, M. Hommel, J. BoisselA phenomenological model of the growth of the necrotic area in ischemic stroke. Mathematical and Computer Modelling, 51 (2010), pp. 1011-1025

[41] T. Heinonen, P. Dastidar, P. Kauppinen, J. Malmivuo, H. EskolaSemi-automatic tool for segmentation and volumetric analysis of medical images. Medical & Biological Engineering & Computing, 36 (1998), pp. 291-296

[42] N. Hevia-Montiel, C. Rosso, N. Chupin, S. Deltour, E. Bardinet, D. Dormont, Y. Samson, S. BailletAutomatic prediction of infarct growth in acute ischemic stroke from MR apparent diffusion coefficient maps. Academic Radiology, 15 (2008), pp. 77-83

[43] S. Huang, Q. Shen, T.Q. DuongArtificial neural network prediction of ischemic tissue fate in acute stroke imaging. Journal of Cerebral Blood Flow and Metabolism, 30 (2010), pp. 1661-1670

[44] M.A. Jacobs, R.A. Knight, H. Soltanian-Zadeh, Z.G. Zheng, A.V. Goussev, D.J. Peck, J.P. Windham, M. ChoppUnsupervised segmentation of multiparameter MRI in experimental cerebral ischemia with comparison to T2, diffusion, and ADC MRI parameters and histopathological validation. Journal of Magnetic Resonance Imaging, 11 (2000), pp. 425-437

[45] M.A. Jacobs, Z.G. Zhang, R.A. Knight, H. Soltanian-Zadeh, A.V. Goussev, D.J. Peck, M. ChoppA model for multiparametric MRI tissue characterization in experimental cerebral ischemia with histological validation in rat: part 1. Stroke, 32 (2001), pp. 943-949

[46] M.A. Jacobs, P. Mitsias, H. Soltanian-Zadeh, S. Santhakumar, A. Ghanei, R. Hammond, D.J. Peck, M. Chopp, S. PatelMultiparametric MRI tissue characterization in clinical stroke with correlation to clinical outcome: part 2. Stroke, 32 (2001), pp. 950-957

[47] J.R. James, K.K. Yoder, O. Osuntokun, A. Kalnin, A. Bruno, E.D. MorrisA supervised method for calculating perfusion/diffusion mismatch volume in acute ischemic stroke. Computers in Biology and Medicine, 36 (2006), pp. 1268-1287

[48] R. Józwiak, A.P. OstrekConceptual improvements in computer-aided diagnosis of acute stroke. Journal of Medical Informatics & Technologies Selected full texts, 17 (2011), pp. 191-199

[49] Y. Kabir, M. Dojat, B. Scherrer, F. Forbes, C. GarbayMultimodal MRI segmentation of ischemic stroke lesions Conference Proceedings IEEE Engineering in Medicine and Biology Society, 2007 (2007), pp. 1595-1598

[50] M.A. Kalafut, D.L. Schriger, J.L. Saver, S. StarkmanDetection of early CT signs of > 1/3 middle cerebral artery infarctions: interrater reliability and sensitivity of CT interpretation by physicians involved in acute stroke care. Stroke, 31 (2000), pp. 1667-1671

[51] Kane, T. Carpenter, F. Chappell, C. Rivers, P. Armitage, P. Sandercock, J. WardlawComparison of 10 different magnetic resonance perfusion imaging processing methods in acute ischemic stroke: effect on lesion size, proportion of patients with diffusion/perfusion mismatch, clinical scores, and radiologic outcomes. Stroke, 38 (2007), pp. 3158-3164

[52] D. Le Bihan, E. Breton, D. Lallemand, P. Grenier, E. Cabanis, M. Laval-JeantetMR imaging of intravoxel incoherent motions: application to diffusion and perfusion in neurologic disorders

[53] Radiology, 161 (1986), pp. 401-407

[54] C.B. Grandin, T.P. Duprez, A.M. Smith, C. Oppenheim, A. Peeters, A.R. Robert, G. CosnardWhich MR-derived perfusion parameters are the best predictors of infarct growth in hyperacute stroke? comparative study between relative and quantitative measurements. Radiology, 223 (2002), pp. 361-370

[55] T. Lelekov-Boissard, G. Chapuisat, J.-P. Boissel, E. Grenier, M.-A. DronneExploration of beneficial and deleterious effects of inflammation in stroke: dynamics of inflammation cells. Philosophical Transactions. Series A, Mathematical, Physical, and Engineering Sciences, 367 (2009), pp. 4699-4716

[56] W. Li, J. Tian, E. Li, J. DaiRobust unsupervised segmentation of infarct lesion from diffusion tensor MR images using multiscale statistical classification and partial volume voxel reclassification. NeuroImage, 23 (2004), pp. 1507-1518

[57] M. Li, L. Ai, H. He, Z. Zheng, B. Lv, W. Li, J. YiSegmentation of infarct in acute ischemic stroke from MR apparent diffusion coefficient and trace-weighted images. Proceedings of SPIE (2009)

[58] H.K. Ma, J.A. Zavala, L. Churilov, J. Ly, P.M. Wright, T.G. Phan, S. Arakawa, S.M. Davis, G.A. DonnanThe hidden mismatch: an explanation for infarct growth without perfusion-weighted

*Characterization of Brain Stroke Using Image and Signal Processing Techniques DOI: http://dx.doi.org/10.5772/intechopen.96288*

imaging/diffusion-weighted imaging mismatch in patients with acute ischemic stroke. Stroke, 42 (2011), pp. 662-668

[59] G. Majno, A. Ames, J. Chiang, R. WrightNo reflow after cerebral ischemia. Lancet, 290 (1967), pp. 569-570

[60] J.A. Maldjian, J. Chalela, S.E. Kasner, D. Liebeskind, J.A. DetreAutomated CT segmentation and analysis for acute middle cerebral artery stroke. AJNR. American Journal of Neuroradiology, 22 (2001), pp. 1050-1055

[61] Martel, S. Allder, G. Delay, P. Morgan, A. Moody Measurement of infarct volume in stroke patients using adaptive segmentation of diffusion weighted MR images. Medical Image Computing and Computer-Assisted Intervention–MICCAI'99. Springer (1999), pp. 22-31

[62] M. Matesin, S. Loncaric, D. PetravicA rule-based approach to stroke lesion analysis from CT brain images. Image and Signal Processing and Analysis, 2001. ISPA 2001. Proceedings of the 2nd International Symposium on. IEEE (2001), pp. 219-223

[63] M. Meilunas, A. Usinskas, R. Kirvaitis, R. DobrovolskisAutomatic contouring of segmented human brain ischemic stroke region on CT images. Mathematical Modelling and Analysis, 8 (2003), pp. 43-50

[64] M. Moseley, J. Kucharczyk, J. Mintorovitch, Y. Cohen, J. Kurhanewicz, N. Derugin, H. Asgari, D. NormanDiffusion-weighted MR imaging of acute stroke: correlation with T2-weighted and magnetic susceptibility-enhanced MR imaging in cats. American Journal of Neuroradiology, 11 (1990), pp. 423-429

[65] K.W. Muir, J. Baird-Gunning, L. Walker, T. Baird, M. McCormick, S.B. CouttsCan the ischemic penumbra be identified on noncontrast CT of acute stroke? Stroke, 38 (2007), pp. 2485-2490

[66] J. Murray. Mathematical Biology, Volume 2, Springer (2002)

[67] D.G. Na, V.N. Thijs, G.W. Albers, M.E. Moseley, M.P. MarksDiffusionweighted MR imaging in acute ischemia: value of apparent diffusion coefficient and signal intensity thresholds in predicting tissue at risk and final infarct size. AJNR. American Journal of Neuroradiology, 25 (2004), pp. 1331-1336

[68] Y. Nagakane, S. Christensen, C. Brekenfeld, H. Ma, L. Churilov, M.W. Parsons, C.R. Levi, K.S. Butcher, A. Peeters, P.A. Barber, C.F. Bladin, D.A. De Silva, J. Fink, T.E. Kimber, D.W. Schultz, K.W. Muir, B.M. Tress, P.M. Desmond, S.M. Davis, G.A. Donnan, EPITHET InvestigatorsEPITHET: positive result after reanalysis using baseline diffusion-weighted imaging/perfusion-weighted imaging co-registration. Stroke, 42 (2011), pp. 59-64

[69] V. Nguyen, H. Pien, N. Menenzes, C. Lopez, C. Melinosky, O. Wu, A. Sorensen, G. Cooperman, H. Ay, W. KoroshetzStroke tissue outcome prediction using a spatially-correlated model. Program and Proceedings of PPIC, 8 (2008), pp. 238-241

[70] P. O'Brien, R.J. Sellar, J.M. WardlawFogging on T2-weighted MR after acute ischaemic stroke: how often might this occur and what are the implications? Neuroradiology, 46 (2004), pp. 635-641

[71] M.V.S.M.L.R.M. Olivot, J.M. Mlynash, G. AlbersOptimal Tmax threshold for predicting penumbral tissue in acute stroke Stroke, 40 (2009), pp. 469-475

[72] C. Oppenheim, Y. Grandin, C. Samson, A. Smith, T. Duprez, C. Marsault, G. CosnardIs there an apparent diffusion coefficient threshold in predicting tissue viability in hyperacute stroke? Stroke, 32 (2001), pp. 2486-2491

[73] J.R. Petrella, J.M. ProvenzaleMR perfusion imaging of the brain: techniques and applications. AJR. American Journal of Roentgenology, 175 (2000), pp. 207-219

[74] D. Pham, C. Xu, J. PrinceCurrent methods in medical image segmentation 1. Annual Review of Biomedical Engineering, 2 (2000), pp. 315-337

[75] T. Phan, G. Donnan, V. Srikanth, J. Chen, D. ReutensHeterogeneity in infarct patterns and clinical outcomes following internal carotid artery occlusion. Archives of Neurology, 66 (2009), p. 1523

[76] K.N. Bhanu Prakash, V. Gupta, M. Bilello, N.J. Beauchamp, W.L. NowinskiIdentification, segmentation, and image property study of acute infarcts in diffusion-weighted images by using a probabilistic neural network and adaptive Gaussian mixture model. Academic Radiology, 13 (2006), pp. 1474-1484

[77] J.-H. Rha, J.L. SaverThe impact of recanalization on ischemic stroke outcome: a meta-analysis. Stroke, 38 (2007), pp. 967-973

[78] C.S. Rivers, J.M. Wardlaw, P.A. Armitage, M.E. Bastin, T.K. Carpenter, V. Cvoro, P.J. Hand, M.S. DennisDo acute diffusion- and perfusionweighted MRI lesions identify final infarct volume in ischemic stroke? Stroke, 37 (2006), pp. 98-104

[79] C.S. Rivers, J.M. Wardlaw, P.A. Armitage, M.E. Bastin, P.J. Hand, M.S. DennisAcute ischemic stroke lesion measurement on diffusion-weighted

imaging—important considerations in designing acute stroke trials with magnetic resonance imaging. Journal of Stroke and Cerebrovascular Diseases, 16 (2007), pp. 64-70

## *Edited by Yongxia Zhou*

This book examines the principles and applications of biomedical imaging and signals processing as well as the advances of multimodal imaging and multi-feature quantification for disease diagnosis and treatments in ophthalmology, stroke, chemotherapy, and neurology. Chapters cover such topics as image segmentation and registration, feature selection for classification, micro-texture characterization, simulation of tissue deformation, and high-level statistical analyses. The chapters also discuss different imaging modalities including MRI and EEG, confocal microscopy, and molecular imaging for improving the accuracy of disease detection via higher spatiotemporal resolution and better illustration. Overall, the book provides a comprehensive review of biomedical imaging and signal processing, informing readers with current and insightful knowledge in these fields.

Published in London, UK © 2021 IntechOpen © olegback / iStock

Biomedical Signal and Image Processing

IntechOpen Book Series

Biomedical Engineering, Volume 10

Biomedical Signal

and Image Processing

*Edited by Yongxia Zhou*