Distribution of prostate cancer for optimized biopsy protocols

Abstract

Prostate cancer is the leading cause of death for American men. The gold standard for diagnosis of prostate cancer is transrectal ultrasound-guided needle core biopsy. Unfortunately, no imaging modality, including ultrasound, can effectively differentiate prostate cancer from normal tissues. As a result, most current prostate needle biopsy procedures have to be performed under empiric protocols, leading to unsatisfactory detection rate. The goal of this research is to establish an accurate 3D distribution map of prostate cancer and develop optimized biopsy protocols. First, we used real prostate specimens with localized cancer to reconstruct 3D prostate models. We then divided each model into zones based on clinical conventions, and calculate cancer presence in each zone. As a result, an accurate 3D prostate cancer distribution map was developed using 281 prostate models. Finally, the linear programming approach was used to optimize biopsy protocols using anatomy and symmetry constraints, and the optimized protocols were developed with various criteria.