Stephanie A Harmon1,2, Thomas H Sanford2,3, G Thomas Brown2,4, Chris Yang1, Sherif Mehralivand1, Joseph M Jacob3, Vladimir A Valera5, Joanna H Shih6, Piyush K Agarwal5, Peter L Choyke1, Baris Turkbey1. 1. Molecular Imaging Branch, National Cancer Institute, Bethesda, MD. 2. Clinical Research Directorate, Frederick National Laboratory for Cancer Research, Frederick, MD. 3. Department of Urology, Upstate Medical University, Syracuse, NY. 4. National Library of Medicine, National Institutes of Health, Bethesda, MD. 5. Urologic Oncology Branch, National Cancer Institute, Bethesda, MD. 6. Division of Cancer Treatment and Diagnosis, Biometric Research Program, National Cancer Institute, Bethesda, MD.
Abstract
PURPOSE: To develop an artificial intelligence (AI)-based model for identifying patients with lymph node (LN) metastasis based on digital evaluation of primary tumors and train the model using cystectomy specimens available from The Cancer Genome Atlas (TCGA) Project; patients from our institution were included for validation of the leave-out test cohort. METHODS: In all, 307 patients were identified for inclusion in the study (TCGA, n = 294; in-house, n = 13). Deep learning models were trained from image patches at 2.5×, 5×, 10×, and 20× magnifications, and spatially resolved prediction maps were combined with microenvironment (lymphocyte infiltration) features to derive a final patient-level AI score (probability of LN metastasis). Training and validation included 219 patients (training, n = 146; validation, n = 73); 89 patients (TCGA, n = 75; in-house, n = 13) were reserved as an independent testing set. Multivariable logistic regression models for predicting LN status based on clinicopathologic features alone and a combined model with AI score were fit to training and validation sets. RESULTS: Several patients were determined to have positive LN metastasis in TCGA (n = 105; 35.7%) and in-house (n = 3; 23.1%) cohorts. A clinicopathologic model that considered using factors such as age, T stage, and lymphovascular invasion demonstrated an area under the curve (AUC) of 0.755 (95% CI, 0.680 to 0.831) in the training and validation cohorts compared with the cross validation of the AI score (likelihood of positive LNs), which achieved an AUC of 0.866 (95% CI, 0.812 to 0.920; P = .021). Performance in the test cohort was similar, with a clinicopathologic model AUC of 0.678 (95% CI, 0.554 to 0.802) and an AI score of 0.784 (95% CI, 0.702 to 0.896; P = .21). In addition, the AI score remained significant after adjusting for clinicopathologic variables (P = 1.08 × 10-9), and the combined model significantly outperformed clinicopathologic features alone in the test cohort with an AUC of 0.807 (95% CI, 0.702 to 0.912; P = .047). CONCLUSION: Patients who are at higher risk of having positive LNs during cystectomy can be identified on primary tumor samples using novel AI-based methodologies applied to digital hematoxylin and eosin-stained slides.
PURPOSE: To develop an artificial intelligence (AI)-based model for identifying patients with lymph node (LN) metastasis based on digital evaluation of primary tumors and train the model using cystectomy specimens available from The Cancer Genome Atlas (TCGA) Project; patients from our institution were included for validation of the leave-out test cohort. METHODS: In all, 307 patients were identified for inclusion in the study (TCGA, n = 294; in-house, n = 13). Deep learning models were trained from image patches at 2.5×, 5×, 10×, and 20× magnifications, and spatially resolved prediction maps were combined with microenvironment (lymphocyte infiltration) features to derive a final patient-level AI score (probability of LN metastasis). Training and validation included 219 patients (training, n = 146; validation, n = 73); 89 patients (TCGA, n = 75; in-house, n = 13) were reserved as an independent testing set. Multivariable logistic regression models for predicting LN status based on clinicopathologic features alone and a combined model with AI score were fit to training and validation sets. RESULTS: Several patients were determined to have positive LN metastasis in TCGA (n = 105; 35.7%) and in-house (n = 3; 23.1%) cohorts. A clinicopathologic model that considered using factors such as age, T stage, and lymphovascular invasion demonstrated an area under the curve (AUC) of 0.755 (95% CI, 0.680 to 0.831) in the training and validation cohorts compared with the cross validation of the AI score (likelihood of positive LNs), which achieved an AUC of 0.866 (95% CI, 0.812 to 0.920; P = .021). Performance in the test cohort was similar, with a clinicopathologic model AUC of 0.678 (95% CI, 0.554 to 0.802) and an AI score of 0.784 (95% CI, 0.702 to 0.896; P = .21). In addition, the AI score remained significant after adjusting for clinicopathologic variables (P = 1.08 × 10-9), and the combined model significantly outperformed clinicopathologic features alone in the test cohort with an AUC of 0.807 (95% CI, 0.702 to 0.912; P = .047). CONCLUSION:Patients who are at higher risk of having positive LNs during cystectomy can be identified on primary tumor samples using novel AI-based methodologies applied to digital hematoxylin and eosin-stained slides.
Authors: Jiayun Li; Karthik V Sarma; King Chung Ho; Arkadiusz Gertych; Beatrice S Knudsen; Corey W Arnold Journal: AMIA Annu Symp Proc Date: 2018-04-16
Authors: Le Hou; Vu Nguyen; Ariel B Kanevsky; Dimitris Samaras; Tahsin M Kurc; Tianhao Zhao; Rajarsi R Gupta; Yi Gao; Wenjin Chen; David Foran; Joel H Saltz Journal: Pattern Recognit Date: 2018-09-13 Impact factor: 7.740
Authors: Ilya G Goldberg; Chris Allan; Jean-Marie Burel; Doug Creager; Andrea Falconi; Harry Hochheiser; Josiah Johnston; Jeff Mellen; Peter K Sorger; Jason R Swedlow Journal: Genome Biol Date: 2005-05-03 Impact factor: 13.583
Authors: Georgios Gakis; Jason Efstathiou; Seth P Lerner; Michael S Cookson; Kirk A Keegan; Khurshid A Guru; William U Shipley; Axel Heidenreich; Mark P Schoenberg; Arthur I Sagaloswky; Mark S Soloway; Arnulf Stenzl Journal: Eur Urol Date: 2012-08-14 Impact factor: 20.096
Authors: Andrea Mari; Riccardo Campi; Riccardo Tellini; Giorgio Gandaglia; Simone Albisinni; Mohammad Abufaraj; Georgios Hatzichristodoulou; Francesco Montorsi; Roland van Velthoven; Marco Carini; Andrea Minervini; Shahrokh F Shariat Journal: World J Urol Date: 2017-11-16 Impact factor: 4.226
Authors: Jianfang Liu; Tara Lichtenberg; Katherine A Hoadley; Laila M Poisson; Alexander J Lazar; Andrew D Cherniack; Albert J Kovatich; Christopher C Benz; Douglas A Levine; Adrian V Lee; Larsson Omberg; Denise M Wolf; Craig D Shriver; Vesteinn Thorsson; Hai Hu Journal: Cell Date: 2018-04-05 Impact factor: 41.582