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

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’s potential as a trustworthy and theoretically grounded solution for high-stakes DTR applications.

In real-world hospitals, doctors often have to make quick decisions using complex medical data, and it’s not always clear which treatment is best—especially when patient conditions are changing rapidly. This paper introduces SAFER, a new system designed to help doctors make safer and more personalized treatment decisions for seriously ill patients, such as those with sepsis. SAFER uses both structured medical records (like lab tests and vitals) and doctors’ notes to learn how different patients respond to treatments. It also knows when it’s uncertain and avoids making risky recommendations. By providing confidence scores and safety checks, SAFER helps reduce the chance of harmful decisions. In tests using real hospital data, SAFER performed better than existing systems and showed it could reduce patient death rates. This work emphasizes the critical responsibility of the research community to ensure safety, ethical standards, andtangible benefits to patient care when advancing such technologies. Therefore, SAFER brings us closer to AI systems that doctors can trust in high-stakes healthcare settings.