Differential Segmentation of the Prostate in MR Images Using Combined 3D Shape Modelling and Voxel Classification

Benign prostatic hyperplasia (BPH) is a non-cancerous expansion of the prostate, the progress of which can be quantified by measuring the relative volumes of the prostate's peripheral zone and central gland. Here we describe a method of automatic segmentation of both regions of the prostate from MR images using a combination of grey-level voxel classification and 3D statistical shape modelling