| Abstrakt: |
A recent study conducted at Nippon Medical School Hospital in Tokyo, Japan, focused on predicting clinically significant prostate cancer using deep learning analysis with multimodal medical data. The study included 178 patients with PSA levels of 20 ng/mL who underwent ultrasound-guided prostate biopsy. The results showed that the deep learning approach had a high predictive accuracy, with an AUC of 0.878 in all patients and 0.862 in patients with PSA levels restricted to 20 ng/mL. This research aims to assist physicians in personalized medical workflows for prostate cancer management. [Extracted from the article] |
