You have accessJournal of UrologyStone Disease: Surgical Therapy VI (MP63)1 Apr 2020MP63-06 PREDICTING THE POSTOPERATIVE OUTCOME OF PERCUTANEOUS NEPHROLITHOTOMY WITH MACHINE LEARNING SYSTEM: SOFTWARE VALIDATION AND COMPARATIVE ANALYSIS WITH GUY’S STONE SCORE AND THE CROES NOMOGRAM Alireza Aminsharifi*, Dariush Irani, Sona Tayebi, Tayebeh Shabanian, and Mehdi Parsaei Alireza Aminsharifi*Alireza Aminsharifi* More articles by this author , Dariush IraniDariush Irani More articles by this author , Sona TayebiSona Tayebi More articles by this author , Tayebeh ShabanianTayebeh Shabanian More articles by this author , and Mehdi ParsaeiMehdi Parsaei More articles by this author View All Author Informationhttps://doi.org/10.1097/JU.0000000000000938.06AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookLinked InTwitterEmail Abstract INTRODUCTION AND OBJECTIVE: To validate the output of a machine learning-based software as an intelligible interface for predicting multiple outcomes after percutaneous nephrolithotomy (PCNL). We compared the performance of this system with Guy’s stone score (GSS) and the Clinical Research Office of Endourological Society (CROES) nomogram. METHODS: Data from 146 adult patients (87 males, 59%) who underwent PCNL at our institute were used, and preoperative variables and postoperative outcomes were recorded. Stone-free status was determined by computed tomography scan postoperatively. Preoperative data were consecutively imported into the software and its output was extracted. To validate the system, accuracy of the software for predicting each postoperative outcome was compared to the actual outcome. Similarly, preoperative data were analyzed with GSS and CROES nomograms to determine stone-free status as predicted by these nomograms, and the predictive performance of these two systems was calculated. A receiver operating characteristic (ROC) curve was generated for each scoring system, and the area under curve (AUC) was calculated and used to assess the predictive performance of the nomograms versus ANN software. RESULTS: Overall stone-free rate (SFR) was 72.6% (106/146). Forty of 146 patients (27.4%) were scheduled for Forty-two ancillary procedures (shock wave lithotripsy (SWL) (n= 31) or repeat PCNL (n=11)) to manage residual renal stones. Overall, the machine learning system predicted the PCNL outcomes with an accuracy ranging between 80% and 95.1%. A higher GSS grade and lower CROES nomogram score were significantly associated with lower SFR (p=0.01 and p= 0.03, respectively). When ROC curves were plotted for each predictive model for stone-free status, the area under the ROC curve for ANN software (0.915) was significantly larger than the AUC for GSS (0.615) or CROES nomograms (0.621) (p<0.001) (Figure 1). CONCLUSIONS: At the internal institutional level, the machine learning-based software was a promising tool for recording, processing and predicting outcomes after PCNL. Validation of this system against an external dataset is highly recommended before its widespread application. Source of Funding: Shiraz University of Medical Sciences. © 2020 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 203Issue Supplement 4April 2020Page: e956-e956 Advertisement Copyright & Permissions© 2020 by American Urological Association Education and Research, Inc.MetricsAuthor Information Alireza Aminsharifi* More articles by this author Dariush Irani More articles by this author Sona Tayebi More articles by this author Tayebeh Shabanian More articles by this author Mehdi Parsaei More articles by this author Expand All Advertisement PDF downloadLoading ...