S Transin1, R Souchon2, C Gonindard-Melodelima3, R de Rozario4, P Walker5, M Funes de la Vega6, R Loffroy7, L Cormier8, O Rouvière9. 1. Department of Radiology, University Hospital François-Mitterrand, 21000 Dijon, France. 2. Inserm, U1032, LabTau, 69003 Lyon, France. 3. CNRS, UMR 5553, Laboratoire d'Ecologie Alpine, Université Grenoble-Alpes, 38041 Grenoble, France. 4. Inserm, U1032, LabTau, 69003 Lyon, France; Département Informatique, Faculté des Sciences et Technologies, Université Lyon 1, 69000 Lyon, France. 5. Medical Imaging Group, Laboratory of Electronics, Computer Science and Imaging (Le2I), CNRS 6306, University of Burgundy, 21000 Dijon, France; Department of MR Spectroscopy, University Hospital François-Mitterrand, 21000 Dijon, France. 6. Department of Pathology, University Hospital François-Mitterrand, 21000 Dijon, France. 7. Department of Radiology, University Hospital François-Mitterrand, 21000 Dijon, France; Medical Imaging Group, Laboratory of Electronics, Computer Science and Imaging (Le2I), CNRS 6306, University of Burgundy, 21000 Dijon, France. 8. Department of Urology, University Hospital François-Mitterrand, 21000 Dijon, France. 9. Inserm, U1032, LabTau, 69003 Lyon, France; Department of Urinary and Vascular Radiology, Hôpital Édouard-Herriot, Hospices Civils de Lyon, 69437 Lyon, France; Université de Lyon, Université Lyon 1, faculté de médecine Lyon Est, Lyon, France. Electronic address: olivier.rouviere@netcourrier.com.
Abstract
PURPOSE: To assess the performance of a computer-aided diagnosis (CADx) system trained at characterizing International Society of Urological Pathology (ISUP) grade≥2 peripheral zone (PZ) prostate cancers on multiparametric magnetic resonance imaging (mpMRI) examinations from a different institution and acquired on different scanners than those used for the training database. PATIENTS AND METHODS: Preoperative mpMRIs of 74 men (median age, 65.7 years) treated by prostatectomy between 2014 and 2017 were retrospectively selected. One radiologist outlined suspicious lesions and scored them using Prostate Imaging-Reporting and Data System version 2 (PI-RADSv2); their CADx score was calculated using a classifier trained on an independent database of 106 patients treated by prostatectomy in another institution. The lesions' nature was assessed by comparison with prostatectomy whole-mounts. Diagnostic accuracy was estimated with areas under receiver operating characteristic curves (AUCs). Sensitivity and specificity were calculated using a CADx threshold (≥0.21) that yielded 95% sensitivity in the training database, and a PI-RADSv2≥3 threshold. RESULTS: A total of 127 lesions (PZ, n=104; transition zone [TZ], n=23) were described. In PZ, CADx and PI-RADSv2 scores had similar AUCs for characterizing ISUP grade≥2 cancers (0.78 [95% confidence interval (CI): 0.69-0.87] vs. 0.74 [95%CI: 0.62-0.82], respectively) (P=0.59). Sensitivity and specificity were respectively 89% (95%CI: 82-97%) and 42% (95%CI: 26-58%) for the CADx score, and 97% (95%CI: 93-100%) and 37% (95%CI: 22-52%) for the PI-RADSv2 score. In TZ, both scores showed poor specificity. CONCLUSION: In this external cohort, the CADx and PI-RADSv2 scores showed similar performances in characterizing ISUP grade≥2 cancers.
PURPOSE: To assess the performance of a computer-aided diagnosis (CADx) system trained at characterizing International Society of Urological Pathology (ISUP) grade≥2 peripheral zone (PZ) prostate cancers on multiparametric magnetic resonance imaging (mpMRI) examinations from a different institution and acquired on different scanners than those used for the training database. PATIENTS AND METHODS: Preoperative mpMRIs of 74 men (median age, 65.7 years) treated by prostatectomy between 2014 and 2017 were retrospectively selected. One radiologist outlined suspicious lesions and scored them using Prostate Imaging-Reporting and Data System version 2 (PI-RADSv2); their CADx score was calculated using a classifier trained on an independent database of 106 patients treated by prostatectomy in another institution. The lesions' nature was assessed by comparison with prostatectomy whole-mounts. Diagnostic accuracy was estimated with areas under receiver operating characteristic curves (AUCs). Sensitivity and specificity were calculated using a CADx threshold (≥0.21) that yielded 95% sensitivity in the training database, and a PI-RADSv2≥3 threshold. RESULTS: A total of 127 lesions (PZ, n=104; transition zone [TZ], n=23) were described. In PZ, CADx and PI-RADSv2 scores had similar AUCs for characterizing ISUP grade≥2 cancers (0.78 [95% confidence interval (CI): 0.69-0.87] vs. 0.74 [95%CI: 0.62-0.82], respectively) (P=0.59). Sensitivity and specificity were respectively 89% (95%CI: 82-97%) and 42% (95%CI: 26-58%) for the CADx score, and 97% (95%CI: 93-100%) and 37% (95%CI: 22-52%) for the PI-RADSv2 score. In TZ, both scores showed poor specificity. CONCLUSION: In this external cohort, the CADx and PI-RADSv2 scores showed similar performances in characterizing ISUP grade≥2 cancers.
Authors: Indrani Bhattacharya; Yash S Khandwala; Sulaiman Vesal; Wei Shao; Qianye Yang; Simon J C Soerensen; Richard E Fan; Pejman Ghanouni; Christian A Kunder; James D Brooks; Yipeng Hu; Mirabela Rusu; Geoffrey A Sonn Journal: Ther Adv Urol Date: 2022-10-10