PERFORMANCE COMPARISON OF STANDARD SEGMENTATION METHODS FOR IDENTIFICATION OF HEMATOMA IN BRAIN CT SCAN IMAGES

In medical image analysis for computer aided diagnosis, segmentation is required as a preliminary step. Segmentation of the brain is very important for detecting tumors, edema, hematoma or any other abnormality but due to complex structure of the brain it becomes a difficult and challenging task. CT scan is the preferred method for traumatic brain injuries because of better contrast on bone, low cost and wide availability. Therefore our research considers the CT scan images. This paper compares the performance of well known segmentation methods on gray level 2D brain CT scan images for detecting hematomas such that most accurate segmentation method can be chosen for further analysis.