Cohort-Anchored Foundation Models for Electronic Health Records: From Risk Scores to Auditable Peer Cohorts

2026-06-20

AI Summary

Existing foundation models for electronic health records (EHRs) are often limited in clinical deployment due to insufficient interpretability, sensitivity to distribution shifts, and misalignment with clinical reasoning. This work proposes the Cohort-Aware Foundation Model (CAFM) framework, which uniquely treats patient cohorts as first-class modeling entities throughout four integrated stages: bias-aware data curation, cohort-conditioned pretraining, multimodal cohort alignment, and clinician-in-the-loop optimization. By design, CAFM yields auditable and trustworthy clinical decision support and can be seamlessly integrated to enhance existing EHR models. The framework’s efficacy is demonstrated across four diverse tasks: acute kidney injury prediction, electrocardiogram-based cardiovascular risk stratification, optic neuropathy triage from orbital imaging, and retinal electrophysiology report generation.

Abstract

Foundation models have achieved remarkable performance across medical question answering, imaging, and electronic health record (EHR) tasks, yet reliable clinical deployment remains challenging due to limited interpretability, vulnerability to distribution shift, and weak alignment with clinician reasoning. We argue that these limitations arise because existing approaches prioritize representation learning while treating patient comparison as an emergent property rather than a primary source of clinical evidence. To address this gap, we propose CAFM, a Cohort-Anchored Foundation Model framework that elevates patient cohorts to a first-class object throughout the learning pipeline. The framework consists of four stages: deviation-aware data curation, cohort-conditioned pretraining, multimodal cohort alignment, and clinician-in-the-loop refinement. Together, these stages improve data quality, organize representations around clinically meaningful cohort structure, preserve modality-specific relationships, and support auditable clinical decision-making. The framework is compositional and can augment existing EHR foundation models without modifying their underlying encoders. We illustrate CAFM through four clinical case studies spanning acute kidney injury prediction, cardiovascular risk stratification from electrocardiograms, optic neuropathy triage from orbital imaging, and electroretinogram-grounded report generation. We further present five empirically testable hypotheses and identify open challenges in data quality, irregular temporality, multimodal learning, distribution shift, and evaluation beyond predictive accuracy. We argue that explicitly anchoring foundation models to patient cohorts provides a principled path toward trustworthy clinical AI.

Problem

• foundation models
• electronic health records
• interpretability
• distribution shift
• clinical reasoning

Innovation

• Cohort-Anchored Foundation Models
• Electronic Health Records
• Interpretable AI
• Multimodal Alignment
• Clinician-in-the-Loop