EyeWorld: A Generative World Model of Ocular State and Dynamics

TL;DR

EyeWorld is a generative model that captures ocular dynamics and multimodal imaging to improve interpretation, prognosis, and robustness in ophthalmic decision-making.

Contribution

It introduces a dynamic, generative world model of the eye that unifies multimodal interpretation and longitudinal prognosis within a single framework.

Findings

Learns a stable latent ocular state across modalities

Enables structure-preserving cross-modality translation

Supports forecasting of disease progression

Abstract

Ophthalmic decision-making depends on subtle lesion-scale cues interpreted across multimodal imaging and over time, yet most medical foundation models remain static and degrade under modality and acquisition shifts. Here we introduce EyeWorld, a generative world model that conceptualizes the eye as a partially observed dynamical system grounded in clinical imaging. EyeWorld learns an observation-stable latent ocular state shared across modalities, unifying fine-grained parsing, structure-preserving cross-modality translation and quality-robust enhancement within a single framework. Longitudinal supervision further enables time-conditioned state transitions, supporting forecasting of clinically meaningful progression while preserving stable anatomy. By moving from static representation learning to explicit dynamical modeling, EyeWorld provides a unified approach to robust multimodal interpretation and prognosis-oriented simulation in medicine.