Researchers have discovered a new machine learning framework that distinguishes low- and high-risk prostate cancer with greater accuracy than ever before.
The framework is designed to help physicians and radiologists, in particular, better define treatment options. for patients with prostate cancer, thus reducing the likelihood of unnecessary clinical intervention.
The team of researchers at the Icahn School of Medicine at Mount Sinai and Keck School of Medicine, University of Southern California (USC), who made the discovery that prostate cancer is one of the leading causes of cancer death, second after lung cancer. Although the latest achievements in prostate cancer studies have saved many lives, tools for objective prediction have so far remained unmet. Currently, the standard methods used to assess the risk of prostate cancer are multi-parametric magnetic resonance imaging (mpMRI), which detects prostatic lesions. ions and a system for reporting and presentation of prostate data, version 2 (PI-RADS v2), a five-point system that classifies lesions found in mpMRI.
Together these tools are designed to predict the probability of clinically significant prostate cancer. However, the assessment of PI-RADS v2 is subjective and does not clearly distinguish between intermediate and malignant levels of cancer (points 3, 4 and 5), which often leads to different interpretations among clinicians.
Assistant of Genetics and Genomics The school, Gaurav Pandey, has said through a rigorous and systematic combination of machine training with radiomics, their goal is to provide radiologists and clinical staff with a reliable predictive tool that can ultimately become more effective and personalized patient care. the publication, Bino Varghese, also said that the way to predict prostate cancer progression with high precision is improving and they believe that their objective framework is much needed.