Mesane karsinomunun bilgisayar temelli derecelendirilmesi

Cancer is one of the most important diseases that ends up with loosing lives, despite many new improvements in medicine. Cancer, which has more than a hundred species, causes uncontrolled division and multiplication of cells and abnormalities in cell shape due to the damage in DNA. The early diagnosis and prognosis of cancer has great importance and will help to reduce loss of lives from this disease. Although many methods are used in the diagnosis and prognosis of cancer, the most effective method is to examine the obtained tissue sample under the microscope by the pathologist after various procedures have been applied. At this stage, pathologists diagnose the cancer by examining the structural properties of the cells. Recently, image processing techniques have been applied to obtained pathological images as well as in other fields of medicine. By using these techniques, the cell locations in pathological images can be determined and the cell characteristics can be extracted from the determined cell locations. In this thesis, visualization of the pathological bladder data is performed and the area can be cropped from these images. An algorithm is designed to determine whether the case is at high grade bladder cancer or low grade bladder cancer using the structural features of the cell nucleus in the region specified by these images. Computer Assisted Diagnosis (CAD) is a system, which is developed to help doctors to diagnose the disease and to decide on the treatment method to be applied. For this purpose, a graphical user interface (GUI) was designed by using the MATLAB software. With the help of the extracted features, it can be decided that loaded sample of the bladder is at high grade or low grade. As a result, developed software allows easy and accurate detection of bladder cancer. As a future work, another study may be performed to identify types of the bladder cancer.