Facilitating 3D Spectroscopic Imaging through Automatic Prostate Localization in MR Images Using Random Walker Segmentation Initialized via Boosted Classifiers

Parmeshwar Khurd, Leo Grady, Kalpitkumar Gajera, Mamadou Diallo, Peter Gall, Martin Requardt, Berthold Kiefer, Clifford Weiss, Ali Kamen

Abstract: Magnetic resonance imaging (MRI) plays a key role in the diagnosis, staging and treatment monitoring for prostate cancer. Automatic prostate localization in T2-weighted MR images could facilitate labor-intensive cancer imaging techniques such as 3D chemical shift MR spectroscopic imaging as well as advanced analysis techniques for diagnosis and treatment monitoring. We present a novel method for automatic segmentation of the prostate gland in MR images. Accurate prostate segmentation in MR imagery poses unique challenges. These include large variations in prostate anatomy and disease, intensity inhomogeneities, and near-field artifacts induced by endorectal coils. Our system meets these challenges with two key components. First is the automatic center detection of the prostate with a boosted classifier trained on intensity-based multi-level Gaussian Mixture Model Expectation Maximization (GMM-EM) segmentations of the raw MR images. The second is the use of a shape model in conjunction with Multi-Label Random Walker (MLRW) to constrain the seeding process within MLRW. Our system has been validated on a large database of non-isotropic T2-TSE (Turbo Spin Echo) and isotropic T2-SPACE (Sampling Perfection with Application Optimized Contrasts) images.