Prostate Gleason Score Detection by Calibrated Machine Learning Classification through Radiomic Features

The Gleason score was originally formulated to represent the heterogeneity of prostate cancer and helps stratify risk patients affected by this tumor. assigning represents an on H&E stain task performed pathologists upon histopathological examination needle biopsies or surgical specimens. In paper, we propose approach focused automatic classification. We exploit a set 18 radiomic features. feature is directly obtainable from segmented magnetic resonance images. build several models considering supervised machine learning techniques, obtaining with RandomForest classification algorithm precision ranging 0.803 0.888 recall 0.873 0.899. Moreover, aim increase never seen instance detection, sigmoid calibration better tune built model.