The overall goal of this project is to create a diagnosis and patient management strategy (called HIT-PIRADS) for prostate cancer (PCa) that will significantly increase positive predictive value (PPV) for clinically significant PCa detection while minimizing unnecessary prostate biopsies and related morbidities. Due to its interpretable nature and bias correction paradigm, the AI system will generate predictions that physicians can trust. One of the immediate outcomes of the system will be a reproducible risk scoring system that can be used in community hospitals and locales without MRI-subspeciality genitourinary trained radiologists to improve the accuracy of prostate imaging nationwide. In the long term, the investigators expect HIT-PIRADS to be widely adopted in clinics and trigger other treatment and prevention strategies to be developed based on HIT-PIRADS.
