Explainable Artificial Intelligence for Prediction of Complete Surgical Cytoreduction in Advanced-Stage Epithelial Ovarian Cancer

Abstract

Complete surgical cytoreduction (R0 resection) is the single most important prognosticator in epithelial ovarian cancer (EOC). Explainable Artificial Intelligence (XAI) could clarify the influence of static and real-time features in the R0 resection prediction. We aimed to develop an AI-based predictive model for the R0 resection outcome, apply a methodology to explain the prediction, and evaluate the interpretability by analysing feature interactions. The retrospective cohort finally assessed 571 consecutive advanced-stage EOC patients who underwent cytoreductive surgery. An eXtreme Gradient Boosting (XGBoost) algorithm was employed to develop the predictive model including mostly patient- and surgery-specific variables. The Shapley Additive explanations (SHAP) framework was used to provide global and local explainability for the predictive model. The XGBoost accurately predicted R0 resection (area under curve [AUC] = 0.866; 95% confidence interval [CI] = 0.8−0.93). We identified “turning points” that increased the probability of complete cytoreduction including Intraoperative Mapping of Ovarian Cancer Score and Peritoneal Carcinomatosis Index < 4 and <5, respectively, followed by Surgical Complexity Score > 4, patient’s age < 60 years, and largest tumour bulk < 5 cm in a surgical environment of optimized infrastructural support. We demonstrated high model accuracy for the R0 resection prediction in EOC patients and provided novel global and local feature explainability that can be used for quality control and internal audit.

Keywords: Explainable Artificial Intelligence; complete cytoreduction; epithelial ovarian cancer.

Figure 1

Prediction performance of the XGBoost model in the test cohort through the Receiver Operator Characteristic (ROC) (AUC = 0.866) plot demonstrating also the CI (AUC = 0.8–0.93). Note that the ROC curves depend on the order of the feature relative risk values without the need to choose a threshold.

Figure 2

A set of beeswarm plots for global explainability of CC0 resection prediction. Dots correspond to the individual EOC patients in the study.

Figure 3

SHAP dependence plots of the top-six global explainability features versus their SHAP value for R0 resection prediction: (A) IMO score, (B) PCI, (C) SCS, and (D) year of surgery.

Figure 4

SHAP Interaction Value Dependence plots: (A) IMO score with type of surgery (B) PCI with time of the procedure (C) PCI with ECOG status (D) Size of largest tumor deposit with SCS.

Figure 5

Succinct visual summary of feature integration by the XAI model into a single risk for R0 resection prediction. The SHAP force plots illustrate examples of explained risks for individual patients. For R0 resection risk, blue features have values that increased the risk, while red features decreased the risk. The combination of impacts of all features is the predicted R0 prediction risk. (A–D) The odds for R0 resection range between 1.54 and 2.35 times higher than expected (E). The odds for incomplete cytoreduction are 3.21 times higher than expected. Each feature impact value represents the change in risk when that feature’s value is known versus unknown. These examples clearly demonstrated the complex interactions between patients and ovarian cancer specific features in a dynamic well supported surgical environment.

Figure 6

Distribution of features for global and local explainability of an AI-based model for prediction of complete cytoreduction in EOC patients. B; baseline, T; transition, E; evaluation.