Physiological World Models for Human State Transitions

2026-08-15

AI Summary

This study addresses the limitation of existing health AI systems in modeling physiological dynamics by proposing an event-conditioned physiological world model. By introducing novel body state transition tokens and structured quality scores, this approach explicitly distinguishes prediction from causal inference while quantifying uncertainty. The project establishes a four-tier capability framework and six benchmark tasks, demonstrating robust reliability under distribution shifts. Furthermore, it enables individualized response prediction, intervention simulation, and safety-bounded planning. Ultimately, this research provides a causally interpretable paradigm for personalized health management and clinical decision-making, overcoming critical gaps in current methodologies regarding dynamic physiological evolution and trustworthy AI deployment in healthcare settings.

Abstract

Continuous multimodal sensing now allows human physiology to be observed throughout daily life rather than only during occasional clinical visits. However, most health artificial intelligence systems are designed to recognize current states, estimate risks or analyse individual biomarkers. They do not directly model how physiological states change in response to real-world events, behaviours, contexts and interventions. Here we propose the Physiological World Model (PWM), an event-conditioned framework for learning these changes at the level of the whole person. We introduce the HumanState Transition Token, a structured, quality-scored unit that connects the physiological state before an event with the event or action, relevant context and intervention information, the physiological trajectory after the event, observed outcomes and data quality. We describe four capability levels, from state representation to bounded intervention planning, together with four data acquisition and validation protocols. We also propose six benchmark tasks covering HumanState representation, forecasting across multiple timescales, individualized response prediction, simulation of alternative interventions, bounded planning and reliability under distribution shift. Together, this framework provides a practical path towards personalized health management, behavioural intervention design and clinician-supervised decision support, while clearly separating prediction from causal inference and making uncertainty, safety, governance and limits of use explicit.

Problem

• Physiological World Models
• Human State Transitions
• Event-conditioned Modeling
• Personalized Health Management

Innovation

• Physiological World Model
• HumanState Transition Token
• Event-conditioned Framework
• Bounded Intervention Planning
• Multimodal Sensing