Search papers, labs, and topics across Lattice.
This review synthesizes the foundational concepts and applications of medical world models, which aim to enhance AI in healthcare by dynamically representing patient states and their evolution in response to clinical interventions. By analyzing 1,455 records and identifying 98 relevant sources, the authors categorize the field into four key capabilities: patient state representation, temporal dynamics modeling, intervention-conditioned simulation, and clinician-supervised planning. Despite promising early evidence for trajectory forecasting and intervention comparison, significant challenges remain in achieving reliable clinical translation due to limitations in data quality and validation processes.
Early evidence suggests that medical world models could revolutionize patient care by enabling dynamic, intervention-responsive AI systems, but critical hurdles remain.
Medical world models offer a framework for extending medical artificial intelligence beyond static prediction by representing evolving patient states and modelling how they change over time and in response to clinical interventions. This Review defines the conceptual boundaries, technical foundations, application domains, and evidence requirements of the field through a structured narrative synthesis with reproducible evidence mapping.We screened 1,455 unique records and assembled a corpus of 98 sources, including 14 studies that met a strict empirical definition of a medical world model. The field is organised around four capabilities: patient state representation, temporal dynamics modelling, intervention-conditioned simulation, and clinician-supervised planning. Evidence spans medical imaging, longitudinal electronic health records, treatment response modelling, physiological and multimodal state modelling, ultrasound and surgical interaction, and population and health-system simulation; clinical digital twins are treated as a cross-cutting integration framework.Current studies provide early evidence of technical feasibility for trajectory forecasting and comparison of candidate interventions, but most remain retrospective, task-specific, or preclinical. The evidence base is further limited by incomplete longitudinal intervention data, inconsistent action semantics, limited causal identifiability, long-horizon error accumulation, inadequate uncertainty estimation, and limited external validation. Clinical translation will therefore depend on precise intervention representations, robust causal and mechanistic grounding, calibrated trajectory-level uncertainty, safety-constrained planning, and prospective multicentre validation against clinically meaningful endpoints.
