Transrectal ultrasound prostate biopsy tracking with efficient and accurate deformation estimation

Prostate cancer is the most frequent cancer for men and the second cause of cancer death. Confirmation of cancer requires transrectal prostate biopsies. Tracking systems for transrectal prostate biopsies make it possible to generate biopsy distribution maps for intra- and post-interventional quality control, 3D visualisation of histological results for treatment planning, and to guide clinicians toward non-ultrasound (US) targets. In this paper, a 3D US based tracking system for fast and accurate estimation of nonlinear prostate tissue deformation is proposed. Linear elasticity and inverse consistency a priori models are used to improve performance. A variant of the SSD image distance measure is proposed for filtering of local intensity shifts, which are frequent in US images. System accuracy was evaluated to 0.83 plusmn 0.54 mm using point fiducial on 295 US volumes from 18 patients.