Medical world models: representing medical states, modelling clinical dynamics and guiding intervention policies

2026-06-15

AI Summary

Current medical AI systems are largely confined to static diagnosis or risk scoring, failing to capture the dynamic progression of diseases or the impact of interventions on patient trajectories. This work proposes a “medical world model” that, for the first time, systematically integrates three core capabilities: patient state representation, clinical dynamics modeling, and intervention-based decision support, within a unified, simulatable, and actionable framework. By synergistically combining foundation models, longitudinal modeling, disease simulation, treatment effect estimation, reinforcement learning, and digital twin technologies, the model enables dynamic prediction and optimization of personalized therapeutic strategies. This advancement represents a significant step toward transforming medical AI from static predictive tools into clinically deployable, dynamic decision-support systems.

Abstract

Medical diagnosis and treatment are dynamic processes in which patient states evolve over time and clinical interventions alter future outcomes. Although current medical AI can detect disease, estimate risk and generate reports, many systems still return static labels or scores, offering limited insight into how illness may progress or how alternative interventions may reshape its trajectory. Medical world models adapt the world-model idea from artificial intelligence to healthcare by learning internal simulators of patient-state dynamics. Their long-term goal is to help clinicians anticipate deterioration, compare treatment-conditioned futures and tailor care to individual patients. Yet relevant work remains scattered across foundation models, longitudinal modelling, disease simulation, treatment-effect estimation, reinforcement learning and digital twins. To bridge this gap, this review outlines a roadmap for advancing medical AI from isolated diagnosis and prediction toward medical world models that simulate disease evolution and support intervention decisions. This roadmap is organized around three coupled capabilities: patient-state construction, clinical dynamics modelling and intervention decision support. Across representative systems, the comparison highlights what each capability contributes and how partial components can be integrated into more mature perception--dynamics--planning systems. Finally, we identify the challenges involved in turning plausible rollouts into clinically useful simulators. Related literature is available at https://github.com/1999kevin/awesome_medical_world_models.

Problem

• medical world models
• clinical dynamics
• intervention policies
• patient-state representation
• disease progression simulation

Innovation

• Medical World Models
• Clinical Dynamics Modelling
• Intervention Decision Support
• Patient-State Representation
• Digital Twins