Survival Concept-Based Learning Models

2025-02-09

AI Summary

To address the lack of interpretable modeling methods for censored data prediction in survival analysis, this paper pioneers the integration of concept learning into the field, proposing two end-to-end trainable, concept-driven survival models: SurvCBM (a concept-bottleneck model) and SurvRCM (a concept-reconstruction model). Methodologically, we innovatively unify the Cox proportional hazards framework with Beran’s nonparametric estimator to establish a dual interpretability mechanism, and introduce concept regularization to enforce semantic consistency among learned concepts. Extensive experiments across multiple medical datasets demonstrate that SurvCBM significantly outperforms state-of-the-art survival models—including CoxPH and DeepSurv—as well as the baseline SurvRCM. It achieves simultaneous improvements in both predictive accuracy and clinical interpretability, empirically validating the substantial gains conferred by concept-based modeling in survival analysis.

Technology Category

Machine Learning: Calibration & Uncertainty QuantificationCognitive Modeling & Cognitive Systems: Conceptual Inference and ReasoningReasoning under Uncertainty: Relational Probabilistic Models

Application Category

User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalizationSemantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMsSearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Concept-based learning enhances prediction accuracy and interpretability by leveraging high-level, human-understandable concepts. However, existing CBL frameworks do not address survival analysis tasks, which involve predicting event times in the presence of censored data -- a common scenario in fields like medicine and reliability analysis. To bridge this gap, we propose two novel models: SurvCBM (Survival Concept-based Bottleneck Model) and SurvRCM (Survival Regularized Concept-based Model), which integrate concept-based learning with survival analysis to handle censored event time data. The models employ the Cox proportional hazards model and the Beran estimator. SurvCBM is based on the architecture of the well-known concept bottleneck model, offering interpretable predictions through concept-based explanations. SurvRCM uses concepts as regularization to enhance accuracy. Both models are trained end-to-end and provide interpretable predictions in terms of concepts. Two interpretability approaches are proposed: one leveraging the linear relationship in the Cox model and another using an instance-based explanation framework with the Beran estimator. Numerical experiments demonstrate that SurvCBM outperforms SurvRCM and traditional survival models, underscoring the importance and advantages of incorporating concept information. The code for the proposed algorithms is publicly available.

Problem

• Enhance survival analysis with concept-based learning
• Handle censored event time data effectively
• Provide interpretable predictions for survival tasks

Innovation

• Survival Concept-based Learning
• Cox Proportional Hazards Model
• Beran Estimator Integration