SAFER: A Calibrated Risk-Aware Multimodal Recommendation Model for Dynamic Treatment Regimes

2025-06-07

AI Summary

In dynamic treatment regime (DTR) recommendation for critical care, key challenges include the trade-off between efficacy optimization and safety assurance, high label uncertainty (e.g., ambiguous mortality labels), and underutilization of heterogeneous multimodal data. To address these, we propose a multimodal DTR model integrating structured electronic health records (EHR) and unstructured clinical notes. We introduce a *fuzzy optimal treatment hypothesis* to mitigate label ambiguity in deceased patients; design a *conformal prediction–based safety-aware decision mechanism* that outputs treatment recommendations with 95% statistical confidence; and develop a *table–language bidirectional collaborative learning architecture* for joint EHR–text representation learning and uncertainty-aware modeling. Evaluated on two public sepsis datasets, our method significantly outperforms state-of-the-art approaches, improving both recommendation accuracy and counterfactual survival rates, while ensuring clinical reliability and interpretability.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Uncertainty Representations
• Natural Language Processing: Safety and Robustness

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Large-scale security measurements

Abstract

Dynamic treatment regimes (DTRs) are critical to precision medicine, optimizing long-term outcomes through personalized, real-time decision-making in evolving clinical contexts, but require careful supervision for unsafe treatment risks. Existing efforts rely primarily on clinician-prescribed gold standards despite the absence of a known optimal strategy, and predominantly using structured EHR data without extracting valuable insights from clinical notes, limiting their reliability for treatment recommendations. In this work, we introduce SAFER, a calibrated risk-aware tabular-language recommendation framework for DTR that integrates both structured EHR and clinical notes, enabling them to learn from each other, and addresses inherent label uncertainty by assuming ambiguous optimal treatment solution for deceased patients. Moreover, SAFER employs conformal prediction to provide statistical guarantees, ensuring safe treatment recommendations while filtering out uncertain predictions. Experiments on two publicly available sepsis datasets demonstrate that SAFER outperforms state-of-the-art baselines across multiple recommendation metrics and counterfactual mortality rate, while offering robust formal assurances. These findings underscore SAFER potential as a trustworthy and theoretically grounded solution for high-stakes DTR applications.

Problem

• Optimizing dynamic treatment regimes for precision medicine with risk awareness
• Integrating structured EHR and clinical notes for reliable treatment recommendations
• Providing statistical safety guarantees for treatment decisions using conformal prediction

Innovation

• Integrates structured EHR and clinical notes
• Uses conformal prediction for safety guarantees
• Addresses label uncertainty in treatments