Transrectal Ultrasonography (TRUS) based systematic needle biopsy of the prostate has been widely used clinically in the diagnosis of prostate carcinoma. Current protocols for prostate biopsy, such as the Sextant Protocol, however, have been proven to be insufficient in cancer detection since these protocols were built without having accurate information on 3D distribution of prostate cancers. In this research, our goal is to optimize prostate biopsy protocols by statistically investigating spatial distributions of prostate cancers. Based on the low-resolution nature of ultrasound imaging and the current clinical conventions, we propose to divide each individual prostate gland into different zones that are can be recognized and accessed by the urologists with ultrasound images during biopsy. By calculating cancer appearance inside each of these zones using a large number of prostate samples, we get the overall distributions of prostate cancers based on which an optimal biopsy protocol can be developed.