Urakami A, Arimura H, Takayama Y, et al. - According to findings, prostate cancer patients could potentially be stratified into low- and high-grade groups (GGs) by applying the proposed approach with multiparametric magnetic resonance (mpMR) radiomics in conjunction with two-dimensional joint histogram computed with dynamic contrast-enhanced (DCE) images.

• Experts identified 101 prostate cancer regions extracted from the MR images of 44 patients and then split them into training (n = 31 with 72 cancer regions) and test datasets (n = 13 with 29 cancer regions).

• Each dataset comprised low-grade tumors (International Society of Urological Pathology [ISUP] GG ≤ 2) and high-grade tumors (ISUP GG ≥ 3).

• Overall 137,970 features comprised mpMR image (16 types of images in four sequences)-based and joint histogram (DCE images at 10 phases)-based features for each cancer region.

• The best combinations of features selected using the least absolute shrinkage and selection operator were employed to define nine signatures (a set of significant features associated with GGs).

• A signature that exhibited the best performance was built using six features derived from the joint histograms, DCE original images, and apparent diffusion coefficient maps.

• For the training and test datasets, the generated areas under the receiver operating characteristic curves were 1.00 and 0.985, respectively.