The assessment of prostate cancer aggressiveness is currently based on Gleason grading of histological samples obtained by TRUS-guided biopsies, which can lead to substantial underestimations due to sampling errors. We previously showed that textural features derived from T2-weighted MRI could potentially serve as a non-invasive biomarker for prostate cancer aggressiveness. The aim of this work was to validate these preliminary results in a multi-center study. We found that the combination of intensity and textural features could distinguish between low/intermediate and high-grade with an accuracy of 71%, which was significantly higher than intensity (60%) or textural features (68%) alone.
This abstract and the presentation materials are available to members only; a login is required.