overview of MRI image processing in the detection of the brain tumor images

overview of MRI image processing in the detection of the brain tumor images

This chapter provides an overview of MRI image processing in the detection of the brain tumor images, and mainly we are doing the contrast enhancement, segmentation and classification in MRI images. The main objective of this thesis is to develop a system to assist brain tumor. This chapter helps us to understand the proposed research work in a better way. Also, it presents the ICA – LDA algorithm with ARHE model-based segmentation, which allows us to segment the tumor and non-tumor cells effectively. The main objective of this research work is to design and develop a method for the detection of the brain tumor in an easy way.

1.2 BRAIN

The brain controls all the necessary vital functions of our body. The most complex organ in the human body is the brain, and it is the part of the central nervous system (CNS). The skull part covers the brain, and the brain is composed of three elongated matters-“gray matter (GM), white matter (WM) and cerebrospinal fluid (CSF).” Cerebrospinal fluid (CSF) is a transparent liquid that encapsulates the brain and the spinal cord and provides many functions to CNS. It provides a barrier against the shocks, which consists of glucose, oxygen, and ions. It is distributed throughout the nervous tissue through the sensory nerves. CSF helps in the removal of waste products away from the nervous tissues.

Figure 1.1 Structure of Brain

Figure 1.1 represents the structure and main parts of the brain. The major parts of the brain are, Cerebrum, cerebellum and brain stem. Cerebrum: It is the largest part of the brain and is composed of right and left hemispheres. It performs higher functions like interpreting touch, vision and hearing, as well as speech, reasoning, emotions, learning, and fine control of movement. Cerebellum: It is located under the cerebrum. Its function is to coordinate muscle movements, maintain posture, and balance. Brainstem: acts as a relay center connecting the cerebrum and cerebellum to the spinal cord. It performs many automatic functions such as breathing, heart rate, body temperature, wake and sleep cycles, digestion, sneezing, coughing, vomiting, and swallowing.

1.2  BRAIN TUMOR

A brain tumor, known as an intracranial tumor, is an abnormal mass of tissue in which cells grow and multiply uncontrollably, visually unchecked by the mechanisms that can control normal cells. More than 150 different brain tumors have been documented, but the two main groups of brain tumors are termed primary and metastatic.

Primary brain tumors include tumors that originate from the tissues of the brain or the brain’s immediate surroundings. Primary tumors are categorized as glial (composed of glial cells) or non-glial (developed on or in the structures of the brain, including nerves, blood vessels and glands) and benign or malignant.

Metastatic brain cancer include tumors that arise any where in the body (such as the breast or lungs) and migrate to the brain, usually through the bloodstream through veins or vessels. Metastatic tumors are considered cancer and are malignant.

Metastatic tumors to the brain affect nearly one in four patients with cancer, or an estimated 150,000 people a year. Up to 40 percent of people with lung cancer will develop metastatic brain tumors. In the past, the outcome for patients diagnosed with these tumors was very poor, with typical survival rates of just several weeks. More sophisticated diagnostic tools, in addition to innovative surgical and radiation approaches, have helped survival rates expand up to years; and also allowed for an improved quality of life for patients following diagnosis.

1.2.1 Types of brain tumors:

A brain tumor, known as an intracranial tumor, is an abnormal mass of tissue in which cells grow and multiply uncontrollably, visually unchecked by the mechanisms that can control normal cells. More than 150 different brain tumors have been documented, but the two main groups of brain tumors are termed primary and metastatic.

Primary brain tumors include tumors that originate from the tissues of the brain or the brain’s immediate surroundings. Primary tumors are categorized as glial (composed of glial cells) or non-glial (developed on or in the structures of the brain, including nerves, blood vessels and glands) and benign or malignant.

Metastatic brain cancer include tumors that arise any where in the body (such as the breast or lungs) and migrate to the brain, usually through the bloodstream through veins or vessels. Metastatic tumors are considered cancer and are malignant.

Metastatic tumors to the brain affect nearly one in four patients with cancer, or an estimated 150,000 people a year. Up to 40 percent of people with lung cancer will develop metastatic brain tumors. In the past, the outcome for patients diagnosed with these tumors was very poor, with typical survival rates of just several weeks. More sophisticated diagnostic tools, in addition to innovative surgical and radiation approaches, have helped survival rates expand up to years; and also allowed for an improved quality of life for patients following diagnosis.

Chordomas  are benign, slow-growing tumors that are most prevalent in people ages 50 to 60. Their most common locations are the base of the skull and the lower portion of the spine. Although these tumors are benign, they may invade the adjacent bone and put pressure on nearby neural tissue. These are rare tumors, contributing to only 0.2 percent of all primary brain tumors.

Figure 1.2 MRI image of chordomas

The figure 1.2 represents the spread of the chordoma in brain

Craniopharingiomas typically are benign, but are difficult tumors to remove because of their location near critical structures deep in the brain. They usually arise from a portion of the pituitary gland (the structure that regulates many hormones in the body), so nearly all patients will require some hormone replacement therapy.

Figure 1.3 MRI image of craniopharyngioma

The figure 1.3 represents the spread of the craniopharyngioma in brain

Gangliocytomas, gangliomas and anaplastic gangliogliomas are rare tumors that include neoplastic nerve cells that are relatively well-differentiated, occurring primarily in young adults.

Figure 1.4 MRI image of gangliomas

The figure 1.4 represents the spread of the gangliomas in brain

Glomus jugulare tumors most frequently are benign and typically are located just under the skull base, at the top of the jugular vein. They are the most common form of glomus tumor. However, glomus tumors, in general, contribute to only 0.6 percent of neoplasms of the head and neck.

Figure 1.5 (a,b) MRI image of Glomus jugulare

The figure 1.5 represents the spread of the Glomus jugulare in brain

Meningiomas are the most common benign intracranial tumors, comprising 10 to 15 percent of all brain neoplasms, although a very small percentage are malignant. These tumors originate from the meninges, the membrane-like structures that surround the brain and spinal cord.

Figure 1.6 MRI image of Meningiomas

The figure 1.6 represents the spread of the Meningiomas in brain

          Pineocytomas are generally benign lesions that arise from the pineal cells, occurring predominantly in adults. They are most often well-defined, noninvasive, homogeneous and slow-growing.

Figure 1.7 MRI image of Pineocytomas

The figure 1.7 represents the spread of the Pineocytomas  in brain

Pituitary adenomas are the most common intracranial tumors after gliomas, meningiomas and schwannomas. The large majority of pituitary adenomas are benign and fairly slow-growing. Even malignant pituitary tumors rarely spread to other parts of the body. Adenomas are by far the most common disease affecting the pituitary. They commonly affect people in their 30s or 40s, although they are diagnosed in children, as well. Most of these tumors can be treated successfully.

Figure 1.8 MRI image of Pineocytomas

The figure 1.8 represents the spread of the Pineocytomas  in brain

Schwannomas are common benign brain tumors in adults. They arise along nerves, comprised of cells that normally provide the “electrical insulation” for the nerve cells. Schwannomas often displace the remainder of the normal nerve instead of invading it. Acoustic neuromas are the most common schwannoma, arising from the eighth cranial nerve, or vestibularcochlear nerve, which travels from the brain to the ear. Although these tumors are benign, they can cause serious complications and even death if they grow and exert pressure on nerves and eventually on the brain. Other locations include the spine and, more rarely, along nerves that go to the limbs.

Figure 1.9 MRI image of Schwannomas

The figure 1.9 represents the spread of the Schwannomas  in brain

1.2.2 Types of malignant brain tumors

Gliomas are the most prevalent type of adult brain tumor, accounting for 78 percent of malignant brain tumors. They arise from the supporting cells of the brain, called the glia. These cells are subdivided into astrocytes, ependymal cells and oligodendroglial cells (or oligos). Glial tumors include the following:

Figure 1.10 MRI image of Schwannomas

The figure 1.10 represents the spread of the Schwannomas  in brain

Astrocytomas are the most common glioma, accounting for about half of all primary brain and spinal cord tumors. Astrocytomas develop from star-shaped glial cells called astrocytes, part of the supportive tissue of the brain. They may occur in many parts of the brain, but most commonly in the cerebrum. People of all ages can develop astrocytomas, but they are more prevalent in adults — particularly middle-aged men. Astrocytomas in the base of the brain are more prevalent in children or younger people and account for the majority of children’s brain tumors. In children, most of these tumors are considered low-grade, while in adults, most are high-grade.

Figure 1.11 MRI image of Astrocytomas

The figure 1.11 represents the spread of the Astrocytomas  in brain

Ependymomas are derived from a neoplastic transformation of the ependymal cells lining the ventricular system and account for two to three percent of all brain tumors. Most are well-defined, but some are not.

Figure 1.12 MRI image of Ependymomas

The figure 1.12 represents the spread of the Ependymomas  in brain

Glioblastoma multiforme (GBM) is the most invasive type of glial tumor. These tumors tend to grow rapidly, spread to other tissue and have a poor prognosis. They may be composed of several different kinds of cells, such as astrocytes and oligodendrocytes. GBM is more common in people ages 50 to 70 and are more prevalent in men than women.

This cancer may comprises of the many types of the cells each can clustered and form like a glomerulus and it can make them as a invasive cancer. This type of stage is called as metastasis. And  this is called as malignant cancer.

Figure 1.13 MRI image of Glioblastoma multiforme

The figure 1.13 represents the spread of the Glioblastoma multiforme  in brain

Medulloblastomas usually arise in the cerebellum, most frequently in children. They are high-grade tumors, but they are usually responsive to radiation and chemotherapy.

Figure 1.14 MRI image of Medulloblastomas

The figure 1.14 represents the spread of the Medulloblastomas in brain

Oligodendrogliomas are derived from the cells that make myelin, which is the insulation  for the brian.

Grade II oligodendrogliomas are low grade tumors. This means the tumor cells grow slowly and invade nearby normal tissue. In many cases, they form years before being diagnosed as no symptoms appear.

Grade III oligodendrogliomas are malignant (cancerous). This means they are fast-growing tumors. They are called anaplastic oligodendriogliomas.

Figure 1.15 MRI image of Oligodendrogliomas

The figure 1.15 represents the spread of the Oligodendrogliomas in brain

1.2.3 Other types of brain tumors

Hemangioblastomas are slow-growing tumors, commonly located in the cerebellum. They originate from blood vessels, can be large in size and often are accompanied by a cyst. These tumors are most common in people ages 40 to 60 and are more prevalent in men than women.

Figure 1.16 MRI image of Hemangioblastomas

The figure 1.16 represents the spread of the Hemangioblastomas in brain

Rhabdoid tumors are rare, highly aggressive tumors that tend to spread throughout the central nervous system. They often appear in multiple sites in the body, especially in the kidneys. They are more prevalent in young children, but also can occur in adults.

Figure 1.17 MRI image of Rhabdoid

The figure 1.17 represents the spread of the Rhabdoid in brain

1.2.4 PEDIATRIC BRAIN TUMORS

Brain tumors in children typically come from different tissues than those affecting adults. Treatments that are fairly well-tolerated by the adult brain (such as radiation therapy) may prevent normal development of a child’s brain, especially in children younger than age five.

According to the Pediatric Brain Tumor Foundation, approximately 4,200 children are diagnosed with a brain tumor in the U.S. Seventy-two percent of children diagnosed with a brain tumor are younger than age 15. Most of these brain tumors grow in the posterior fossa (or back) of the brain. Children often present with hydrocephalus (fluid build up in the brain) or the face or body not working properly.

Some types of brain tumors are more common in children than in adults. The most common types of pediatric tumors are medulloblastomas, low-grade astrocytomas (pilocytic), ependymomas, craniopharyngiomas and brainstem gliomas.

The World Health Organization (WHO) had developed a grading system to indicate a tumor’s malignancy or benignity based on its histological features under a microscope.

• Most malignant
• Rapid growth, aggressive
• Widely infiltrative
• Rapid recurrence
• Necrosisprone

1.3 IMAGING OF BRAIN

Brain imaging techniques help radiologists and researchers to detect problems in the human brain, without need of neurosurgery. There are number of techniques available in hospitals throughout the world that are proved to be safe.

(A) MRI Scan In the last few years, the uses of Magnetic Resonance Imaging (MRI) scanners in medical field have grown . To diagnosis a brain tumor, Magnetic Resonance Imaging is commonly used, because of its advantages of high resolution to delicate tissues of brain and it is a non-ionizing, non- radioactive which will not damages human organs. Combined with medical knowledge and clinical experience, the experienced diagnosist can locate the tumor sizes, shapes, anatomical structure and other pathological characteristics of brain tumors which help in suggesting the proper treatment to the patients. Because there are several MRI examinations for every patient in the whole therapeutic treatment, each of which can give data in multiple sequences, it is a large amount of data to be dealt with for the doctors. Long time of hard work will inevitably lead to mistakes in the diagnosis of the tumor contours for the doctors.

                                      Figure 1.18  process of MRI

The figure 1.18 represents the process of magnetic resonance imaging.

Brain tumor detection using MRI scan images is popularly used in the biomedical field for identification and visualization of finer details of the interior parts of the body. This technique is significantly used for detecting the differences between the tissues and resulted to be a better technique when compared with computed tomography (CT). Thus MRI technique is specially used for the detection and identification of brain tumor and cancer imaging. Ionizing radiations are used for CT scan and magnetization is used for MRI scan, here a strong magnetic field is produced to adjust the nuclear magnetization and then uses radio frequencies for the coordinate of the magnetization which are identified by the image scanner. That signal generated is further handled to extract the information of the body. By comparing both the types of images MR image is safer than CT scan image because it is harmless for human body. In the existing scenarios, the radiologist used for brain tumor detection need to manually observe the MRI images and tries to detect and locate the abnormalities present in the brain image. In order to locate the abnormalities consumes time and need lot of efforts. So, there requires an assistant tool which helps in detecting the presence of tumor in the MRI image of brain and stage accurately. Thus detection of tumor in brain plays a curial and tough job in the field of medical image processing. The separation of damaged or infected part from the brain along with its shape, size and boundary is known as detection of brain tumor

(B) CT Scan: Computed Tomography (CT) is used to construct brain images through a series of X-rays scan. In the process of CT scan the patient lies on a table which can slide in to center of CT scanner. Machine X-ray beam rotates around the patient. Scan usually only takes 30 minutes. It shows structural view of brain not the functional.

(C) PET Scan: A Positron Emission Tomography (PET) scan is used to get functional view of brain. A small amount of radioactive material is injected in the body and scanner detects material when this material breaks down. A PET scan is able to show how different areas of the brain behave during various tasks. PET is able to detect following brain abnormalities: – Tumors – Memory disorders – Seizures

Doctors may use MRI scans in diagnosing brain tumor and cancer. Brain tumor using MRI scan images is popularly used in the biomedical field for identification and visualization of finer details of the interior parts of the body. This technique is significantly used for detecting the differences between the tissues and resulted to be a better technique when compared with computed tomography (CT). Thus MRI technique is specially used for the detection and identification of brain tumor and cancer imaging. Ionizing radiations are used for CT scan and magnetization is used for MRI scan, here a strong magnetic field is produced to adjust the nuclear magnetization and then uses radio frequencies for the coordinate of the magnetization which are identified by the image scanner. That signal generated is further handled to extract the information of the body. By comparing both the types of images MR image is safer than CT scan image because it is harmless for human body. In the existing scenarios, the radiologist used for brain tumor detection need to manually observe the MRI images and tries to detect and locate the abnormalities present in the brain image. In order to locate the abnormalities consumes time and need lot of efforts. So, there requires an assistant tool which helps in detecting the presence of tumor in the MRI image of brain and stage accurately. Thus detection of tumor in brain plays a curial and tough job in the field of medical image processing. The separation of damaged or infected part from the brain along with its shape, size and boundary is known as detection of brain tumor. Magnetic Resonance Imaging and Brain Tumors Magnetic-Resonance-Imaging (MRI) is a best visualization method which allows images of internal anatomical structure to be collected in a safe and non-invasive without harming to the human body. It‟s based on the standards of Nuclear Magnetic Resonance which allows immense array of different types of visualizations to be performed. This process of imaging has a particular importance for producing images of the brain, due to the ability of MRI to produce signals that can identify the difference between various soft tissues (such as gray matter and white matter). In medical image processing there are two MRI most often used for visualizations those are T1-weighted and T2-weighted images. These MR slice weightings refer to the important signal (whether it be the T1 time or the T2 time) measured to make the contrast in the image. The areas with high fat content have a short T1 time relative to water; T1-weighted images can be taken as visualizing locations of fat. In contrast areas with high water content have a short T2 time related to areas of high fat content, T2-weighted images can be observed as visualizing locations of water. An example T1 and T2-weighted images and the locations of two normal tissue types in these forms. In finding out brain tumors, a second T1-weighted image is collected frequently after the injection of a „contrast agent . These  contrast agent  components use to have elements whose composition makes a decrease in the T1 time of nearby tissue (gadolinium is one example). The image left: T1-weighted image (light regions visualize locations of fat). The Image Top right: T2-weighted image (light regions visualize locations of water). Below left White matter (high fat) locations.

1.4 IMAGE PROCESSING

Image processing is a methodology which is capable of converting an image into discrete form and it performs certain operations on image, so as to achieve an enhanced image or to extract some vital information from it. It is similar to digital signal processing. In image processing, input is an image (may be a video frame or a photograph in any format) and the output may be an image or the characteristics of the input image. Image processing system usually considers an image as a two dimensional signal, while processing. It is one among the emerging technologies, with its branches of application wide spread into several domains of business. Image processing is a core research area in engineering and it also acts as a thrust area in other disciplines of computer science. Researchers are in need of image processing; as it offers real time applications and the results derived from image processing techniques are also made available to the hands of its user. Generally, medical images obtained from hospitals are in DICOM- Digital Imaging and Communications in Medicine format. These image formats are quite large in size and require higher memory space for storage. For portability of these data, they are converted into JPEG, JPG, BMP, TIFF, PNG file formats. Analysis of images in DICOM format is a tedious process and the images are converted into any of the above said file format, and they are used worldwide. Image processing is extended to such medical image diagnosis so as to identify the pathologies present in our body, especially the pathologies present in human brain that are difficult to diagnose. The work focuses upon this particular issue and is organized to resolve it. Tumor is an uncontrolled growth of tissue in any part of the body. The tumor is of different types and they have different characteristics and different treatment. This work concentrates on developing a robust algorithm for detection of range and shape of tumor in brain MR Images. For brain tumor detection, image segmentation and  is required. This is considered to be one of the most important but difficult part of the process of detecting brain tumor. Hence, it is highly necessary that segmentation of the Magnetic Resonance images must be done accurately, so that further diagnosis is done.

1.4.1 Techniques in image processing:

Image processing is any form of signal processing for which the input is an image, such as a photograph or video frame, the output of image processing may be either an image or a set of characteristics or parameters related to the image. In a general way it can be defined as processing of the given input image to full fill the requirement in the application. Most image-processing techniques involve treating the image as a two- dimensional signal and applying standard signal-processing techniques to it. Image Processing is used in various applications such as, remote Sensing, Medical Imaging, Non-destructive Evaluation, Forensic Studies, Textiles Material Science, Military and Film

1. Analog image processing– Overall the image processing system is divided into two categories, Digital image processing and Analog Image processing In electrical engineering and computer science.

Analogue image processing is an image processing task is conducted on two-dimensional signals by analogue means. Analog Image Processing refers to the alteration of image through electrical means. The most common example is the television image. The television signal is a voltage level which varies in amplitude to represent brightness through the image. By electrically varying the signal, the displayed image appearance is altered. The brightness and contrast controls on a TV set serve to adjust the amplitude and reference of the video signal, resulting in the brightening, darkening and alteration of the brightness range of the displayed image. Analog image can be mathematically represented as a continuous range of values  representing position and intensity. For an analog image the intensity can be represented on a normalized scale from 0 to 1 with infinite divisions, and spatially with an infinite number of coordinates .

2. Digital Image Processing

Digital image processing refers to processing of digital images by means of a digital computer by applying some function to change the value of the pixel to obtain a different function value. Modern digital technology has made it possible to manipulate multi-dimensional signals with systems that range from simple digital circuits to advanced parallel computers.

The goal of this manipulation can be divided into three categories:

• Image Processing – image in image out
• Image Analysis – image in measurements out
• Image Understanding – image in high-level description out

An image defined in the “real world” is considered to be a function of two real variables, for with ’a’ as the amplitude (e.g. brightness) of the image at the real coordinate position .An image may be considered to contain sub-images sometimes referred to as regions–of–interest, ROIs, or simply regions. This concept reflects the fact that images frequently contain collections of objects each of which can be the basis for a region. In a sophisticated image processing system it should be possible to apply specific image processing operations to selected regions. Thus one part of an image (region) might be processed to suppress motion blur while another part might be processed to improve colour rendition. An image may be defined as a two-dimensional function, f(x, y), where x and y are spatial coordinates, and the amplitude of at any pair of coordinates (x, y) is called the intensity or gray level of the image at that point. When x, y and the amplitude values of f are all finite, discrete quantities, the image is called as a digital image. The field of digital image processing refers to processing digital images by means of a digital computer. It should be noted that a digital image is composed of a finite number of elements, each of which has a particular location and value. These elements are referredto as picture 5 elements, image elements, and pixels. Pixel is the term most widely used to denote the elements of a digital image .These definitions are considered formally in the most advanced of the senses, so it is not surprising that images play the single most important role in human perception. However, unlike humans, who are limited to the visual band of the electromagnetic (EM) spectrum, imaging machines cover almost the entire EM spectrum, ranging from gamma to radio waves. They can operate also on images generated by sources that humans are not accustomed to associating with images. These include ultrasound, electron microscopy, and computer-generated images. Thus, digital image processing encompasses a wide and varied field of applications. The amplitudes of a given image will almost always be either real numbers or integer numbers. The latter is usually a result of a quantization process that converts a continuous range (say, between 0 and 100%) to a discrete number of levels. In certain image-forming processes, however, the signal may involve photon counting which implies that the amplitude would be inherently quantized. In other image forming procedures, such as magnetic resonance imaging, the direct physical measurement yields a complex number in the form of a real magnitude and a real phase.

Digital image processing tool deals with manipulation of digital images through a digital computer. It is a subfield of signals and systems but focus as particularly on images. DIP focuses on developing a computer system that is able to perform processing on an image. The input of that system is a digital image and the system processes that image using efficient algorithms and gives an image as an output  digital Images .A digital image is an image which has been digitized both in spatial coordinates and in brightness. We may consider a digital image as a matrix whose row and column indices identify a point in the image whose corresponding matrix elemental value identifies the gray level at that point. The elements of such a digital array are called image elements, picture elements, pixels or pels .This is the definition of a gray level image, also called a gray scale image. The images used in this research are gray scale images.

Gray scale images can be used to show variances in relative intensity for a given scene or subject matter. Because of  the  intensities captured on a MRI are records of the relative absorption of radiation, gray scale images are entirely suitable for digital mammogram images. The elements in a digital image contain a discrete value, usually a positive integer within a given range. Typically images will be defined by the range of values they contain. For example, an eight-bit gray scale image is one in which the pixel values range from 0 to 255. A twelve-bit gray scale image contains pixel values ranging from 0 to 4095. Likewise, a binary, or one-bit, image contains pixels which have values of zero or one.

A disease is a particular abnormal condition, a disorder of a structure or function, that affects part of organism or all of an organism. The causal study of disease is called pathology. Disease is often construed as a medical condition associated with specific symptoms and signs. It may be caused by factors originally from an external source, such as infectious disease, or it may be caused by internal dysfunctions, such as autoimmune diseases. In humans, “disease” is often used more broadly to refer to any condition that causes pain, dysfunction, distress, social problems, or death to the person afflicted, or similar problems for those in contact with the person. In a broader sense, it sometimes includes injuries, disabilities, syndromes, infections, isolated symptoms, deviant behaviors and a typical variations of structure and function, while in other contexts and for other purposes these may be considered distinguishable categories. Diseases usually affect people not only physically, but also emotionally, as contracting and living with a disease can alter one’s perspective on life, and one’s personality.

1.4.2 Steps in image processing:

                   Moreover, it is subjective for the doctors to determine the state of the diseases according to their medical knowledge and clinical experiences. Therefore, developing an automatic or a semi-automatic computer-aided diagnosis system is meaningful in real medical treatments, which can release the workload of doctors and improve the accuracy by giving objective results. This is an important matter in the research field of medical imaging and a lot of algorithms have been suggested to solve it. But unfortunately it is still unsolved due to the low efficiency, accuracy, applicability and robustness of present algorithms. So in this work we proposed a novel brain tumor detection and classification framework from tumor region segmentation to the tissue classification. By adaptive training, the system can obtain the properties of tumors after the first detection and classification and then separate the tumors in the subsequent MRI examinations automatically. Democracy Engineering is indicating how the system is created, designed and implemented politically so that the system is meeting the customers need. The process consists of five modules as represented in figure 1.19 that are  1.Image Enhancement (Image pre-processing) 2. Image Segmentation 3. Feature extraction, 4.segmentation and 5. Classification

• Image preprocessing: to improve the image in ways that increases the chances for success of the other processes.
• Image segmentation: to partition an input image into its constituent parts or objects.
• Feature Estimation: to convert the input data to a form suitable for computer processing and to extract features that result in some quantitative information of interest or features that are basic for differentiating one class of objects from another.
• Data Analysis: to assign a label to an object based on the information provided by its descriptors and to assign meaning to an ensemble of recognized objects. Knowledge about a problem domain is coded into an image processing system in the form of a knowledge database.

Figure 1.19 general process of image processing

1.5 RESEARCH OBJECTIVE

Gap Analyses 

Traditional algorithms are very effective to the initial cluster size and cluster centers. If these clusters varies with different initial input then it create problem in classifying pixels.  In the existing popular Fuzzy algorithm, the cluster centroids values taken randomly. This will increases time to get desired solution.  Currently individual image is used to detect or diagnose patient parameters. But single image may provide less information. Then the images are not classified using their own feature instead user interpretation of image. Hence a novel Convolutional neural network for classification and ICA-LDA algorithm for the process of segmentation.

1.6 HYPOTHESIS

Each MRI image contains information that differs to one another. In proposed method, we are using fusion technique that helps to combine the information of multiple images and hence more information we can gather and also use of two types of texture features gray level run length matrix with ICA-LDA algorithm to get more accurate result.

 1.7 OBJECTIVE

The system should be able to process MRI multi slice sequences, it should be able to obtain tumor region with precise boundary from the pre-processed enhanced image. The image enhancement can be done using the fuzzy and spline based histogram . Then  the region should be segmented to extract the texture features using the ICA – LDA algorithm. In the final stage convolutional CNN classifier was used. Classifier should compare and identify the tumor with high accuracy . Finally it should classify the tumor region as malignant or benign

1.8 MOTIVATION

Radiologist helps manual discrimination of tumor area and provides the details to neurologist. But this task is monotonous due to the enormous volume of images and it requires more time, strength and validity. This motivates us to propose the system which gives accurate calculation, judgement of tumor size, which is of great significance for clinical reasons, like treatment planning and various other therapies.

1.9 METHODOLOGY

Proposed method consists of two phases testing phase and training phase. In training phase, initially pre-processing of input images is done. Features of pre-processed images are extracted. These features are used to train the classifier and resulting coefficients are stored in knowledge base. In testing phase MRI multi sequence images are taken and registration of each image is done before pre-processing. After pre-processing, region of interest is manually selected and feature is extracted as in training phase. Output of which is compared with trained co-efficient stored in knowledge base using a classifier to get segmented classified output. Decision based on type of tumor is also obtained.

1.10 THESIS ORGANIZATION

The thesis was organized in 7 Chapters:

CHAPTER 1: Introduction to brain tumor and magnetic resonance image, system to detect and classify brain tumor using MRI, general categorization of steps involved in brain tumor detection and classification, description of parameters used to get accurate results, understanding the gap bridge, motivation to select this research area, major research contribution and methodology.

CHAPTER 2: Provides brief overview of existing literature on brain tumor detection and classification, describes the various techniques and methodology used to get more accurate result with different algorithms. Finally the chapter ends with summarizing advantages, limitations and challenges of the literature presented on brain tumor detection and classification.

CHAPTER 3: Forms the main part of the dissertation as it introduces the proposed block diagram and methodology of our research work, this chapter also explain the process involved in image enhancement with respected algorithms.

CHAPTER 4: Chapter explains in detail the method of extracting the features from the segmented image by implementing the novel ICA-LDA algorithm.

CHAPTER 5: Describes the implementation and complete work flow about the classification of tumor using CNN classifier, First this section explains the proposed CNN algorithm which provide image for classification and then it explains the efficient method for forming a boundary for effected tissues, then the feature extraction.

CHAPTER 6: chapter provides the different set of results from the proposed system; the results are compared with different existing algorithms. Chapter also provides the bar graph to understand the comparative results.

CHAPTER 7: Draws the conclusion reached and offers the suggestions for future work.

1.11 CONTRIBUTION

The thesis addresses the importance of image enhancement before segmentation and classification. The proposed algorithm for image enhancement, classification, segmentation will produces more accurate result with very fewer features for tumor classification. The proposed system uses the Convolutional CNN to classify the tumor into benign, malignant or normal brain. System mainly proposes the fusion technique to gather more information from the image and reducing the number of features for classification. The system computational time and complexity reduces and increases the accuracy.

1.12 SUMMARY:

This chapter presents about the introduction about image processing and brain tumor . This chapter implicitly explains the need of early detection of the tumor by cost and time effective technology. From this chapter anyone can read and understand the functionalities of the entire thesis concept in a better manner.