You have accessJournal of UrologyImaging/Radiology: Uroradiology I1 Apr 2018MP14-10 PROSTATE CANCER HETEROGENEITY: TEXTURE ANALYSIS OF MULTIPLE MRI SEQUENCES FOR DETECTION AND SELECTION OF BIOPSY TARGETS Clement Orczyk, Arnauld Villers, Henry Rusinek, Vincent Le Pennec, Celine Bazille, Artem Mikheev, Myriam Bernaudin, Audrey Fohlen, and Samuel Valable Clement OrczykClement Orczyk More articles by this author , Arnauld VillersArnauld Villers More articles by this author , Henry RusinekHenry Rusinek More articles by this author , Vincent Le PennecVincent Le Pennec More articles by this author , Celine BazilleCeline Bazille More articles by this author , Artem MikheevArtem Mikheev More articles by this author , Myriam BernaudinMyriam Bernaudin More articles by this author , Audrey FohlenAudrey Fohlen More articles by this author , and Samuel ValableSamuel Valable More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2018.02.503AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES In oncology, heterogeneity of cancer is a driver to progression. This feature can be captured quantitatively by image analysis, especially MRI. Prostate mpMRI already showed capabilities in prostate cancer diagnosis pathway. To develop and test in a biopsy population a texture analysis score, entropy score, capturing heterogeneity, based on analysis of multiple MRI sequences for detection and stratification of prostate cancer in view of selection of targets for biopsy. METHODS Under ethical approval, 134 Volume of Interest (VOIs) were generated from 20 consecutive patients with a clinical 1.5T mpMRI (T2WI, ADC map and DCE WI) at time of biopsy. VOIs comprised zonal anatomy segmentation, biopsy targets and cancer. Reference test was targeted and systematic saturation biopsy. Calibrated Entropy (E) was computed and plotted as a multiparametric score defined as Entropy Score (ES)= E ADC+ E Ktrans + E Ve+ E T2WI. Performances were assessed for detection of significant prostate cancer (SPCa). RESULTS Cancer (GS 6- 8) was found in 12 of the 20 patients with a median PSA of 8.22ng/ml. ES performed respectively an Area Under the Curve of 0.89 and 0.88 for detection of SPCa among the targets and cancer and all VOIs. Best ES estimated threshold of 16.61 NAT led to a sensitivity of 100% and negative predictive value of 100%. SPca (ES=17.96 ±0.72 NAT; CI 95%) showed a significant higher ES than non-SPCa (ES=15.33 ±0.76 NAT). ES correlated with Gleason Score (rs =0.5683, p=0.033) and maximum cancer core length (? = 0.781; p=0.0009). CONCLUSIONS Capturing heterogeneity of PCa across multiple MRI sequences with ES performed high performances for detection and stratification of SPca in this biopsy population with potential to select accurately targets for biopsy. © 2018FiguresReferencesRelatedDetails Volume 199Issue 4SApril 2018Page: e183 Advertisement Copyright & Permissions© 2018MetricsAuthor Information Clement Orczyk More articles by this author Arnauld Villers More articles by this author Henry Rusinek More articles by this author Vincent Le Pennec More articles by this author Celine Bazille More articles by this author Artem Mikheev More articles by this author Myriam Bernaudin More articles by this author Audrey Fohlen More articles by this author Samuel Valable More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...