Diagnosis and Localization of Prostate Cancer via Automated Multiparametric MRI Equipped with Artificial Intelligence

Prostate MRI scans for pre-biopsied patients are important. However, fewer radiologists available diagnoses, which requires multi-sequential interpretations of multi-slice images. To reduce such a burden, artificial intelligence (AI)-based, computer-aided diagnosis is expected to be critical technology. We present an AI-based method pinpointing prostate cancer location and determining tumor morphology using multiparametric MRI. The study enrolled 15 who underwent radical prostatectomy between April 2008 August 2017 at our institution. labeled the area on peripheral zone MR images, comparing with histopathological mapping specimens. Likelihood maps were drawn, tumors divided into morphologically distinct regions superpixel method. consisted pixels, utilize likelihood value computed from T2-weighted, apparent diffusion coefficient, diffusion-weighted MRI-based texture features. Cancer was determined based maps. evaluated diagnostic performance by under receiver operating characteristic (ROC) curve according Chi-square test. ROC 0.985. Sensitivity specificity approach 0.875 0.961 (p < 0.01), respectively. Our procedures successfully applied automated localization shape estimation