The second version of the Prostate Imaging Reporting and Data System (PIRADSv2) indicates the likelihood of a clinically significant cancer with a simplified 5-point scale. To assist radiologists in making diagnostic decisions consistent with the PIRADSv2, we proposed a machine learning-based computer aided diagnosis (CAD) scoring tool of prostate cancer risk evaluation by combining apparent diffusion coefficient (ADC) and T2-weighted MRI-based features. The tool could provide a malignancy prediction color map of 5 scores. The statistical results of the total score test for 130 patients between radiologist graded and the CAD tool showed high accuracy and AUC.
This abstract and the presentation materials are available to members only; a login is required.