[ad_1]
Researchers at the Mount Sinai Icahn School of Medicine and the Keck School of Medicine at the University of Southern California have developed a new machine learning framework that accurately differentiates prostate cancer from prostate cancer. low and high risk. The framework aims to enable physicians, particularly radiologists, to determine treatment options for prostate cancer patients, thereby reducing potentially unnecessary clinical interventions.
Prostate cancer is a leading cause of cancer death in American men, just behind lung cancer. Although advances in prostate cancer research have saved many lives, objective prediction tools remain an unmet need.
Current methods of badessing prostate cancer risk are multiparametric magnetic resonance imaging (MRI) and the Prostate Imaging and Reporting System, version 2 (PI-RADS v2 ), which ranks the lesions found on the pMRI. However, the PI-RADS v2 rating is subjective and makes no clear distinction between intermediate and malignant cancer levels, leading to contrasting interpretations by clinicians.
The combination of machine learning and radiomics has already been suggested, but other studies have only tested a small number of methods. However, by developing a predictive framework to rigorously and consistently evaluate many of these methods, researchers at Mount Sinai and USC were able to identify the most effective method. They were also able to clbadify prostate cancer patients with high sensitivity and even higher predictive value.
EurekAlert!, February 7, 2019
Source link
