Automatic Brain Tissue Detection of Acute Ischemic Stroke Patients

Abstract

Currently, the different algorithms for detecting tumor range and shape in brain MR images are being implemented and it is now possible to find out the degree of tumor with regard to the given tumor area. The information was gathered via research of various statistical analysis methods which are all based on those individuals who have been diagnosed with brain tumors, and then risk factors and symptoms that appear for all individuals diagnosed with brain tumors were discovered. The advancement of research in medicine day and night aims to provide modern therapeutic approaches. The surgeon physically examines this image in order to identify and diagnose brain tumors. However, this procedure accurately measures the stage and scale of the tumor and accurately distinguishes the stage of the tumor based on the location of the tumor. This dissertation employs k-means and fuzzy c-means algorithms to segment brain tumors and classify tumor cells using CNN (convolution neural network). This approach enables the accurate and reproducible segmentation of tumor tissue equal to manual segmentation. Additionally, it decreases research time and accurately determines the stage of tumor from a given region of tumor.