According to a study published in Radiology, a deep learning model has been developed that performs at the level of an experienced abdominal radiologist in detecting clinically significant prostate cancer on MRI. The researchers hope this AI model can serve as an adjunct to radiologists, enhancing the accuracy of prostate cancer detection. Prostate cancer is the second most common cancer in men worldwide, and radiologists typically use multiparametric MRI, interpreted through the Prostate Imaging-Reporting and Data System (PI-RADS), to diagnose it.
However, the PI-RADS system has limitations, especially in lesion classification. Traditional AI approaches require lesions to be annotated by radiologists, which is time-consuming and often lacks consistency. To address this, the researchers developed a new deep-learning model that predicts the presence of clinically significant prostate…
