Early diagnosis represents one of the most promising approaches to reducing the growing cancer burden. In this paper, computer aided diagnosis (CAD) system for detection of Endometrium Cancer is developed by analyzing the ultrasound images of the uterine. CAD system helps to improve the diagnostic performance of radiologists in their image interpretations. The results of every module are demonstrated in this paper. The ultrasound images that are acquired will be filtered using median filter & then segmented to obtain Region of Interest ROI using Region Growing Segmentation Algorithm. The different statistical features of the ROI will then be extracted which acts as an input to the ANN thus providing results as to Cancerous or Non-Cancerous for the tumor.