Fig. 6From: A whole slide image-based machine learning approach to predict ductal carcinoma in situ (DCIS) recurrence riskValidation of 8-feature DCIS recurrence risk prediction model in an independent validation cohort. a Kaplan-Meier curves showing a robust stratification of patients in the validation cohort into high risk of recurrence and low risk of recurrence subgroups. Significance was measured using the log-rank test, and the gray line represents the unstratified full validation cohort. b Univariate and multivariate Cox regression analysis of the validation cohort comparing the influence of common clinicopathological variables on the recurrence risk predictive 8-feature model, for 10-year recurrence-free survivalBack to article page