Prostate Segmentation in Ultrasound Images Using Hybrid Method

Prostate segmentation in ultrasound images is a challenging problem due to the inherent speckle and some artifacts like shadows, attenuation and signal dropout. In this paper we propose an approach, which includes pre-processing for image enhancement and uses hybrid method, which combines a priori boundary, shape and region information of the prostate to improve the segmentation accuracy. It is based on the active contour model and uses a template matching method to define the initial contour at which the segmentation begins. Some experimental results are presented, obtained by computer simulation in the MATLAB environment. Implementation results demonstrate the effectiveness of the approach for application in screening diagnostics.