ORBIT: A Prognostic World Model for Ocular Reasoning Based on Imagined Trajectories

The longitudinal management of blinding fundus diseases constitutes a Partially Observable Markov Decision Process (POMDP) necessitating a critical precision-risk trade-off between intervention and over-treatment, as true pathology is often obscured in static observations. However, existing paradigms fail to address this complexity. Traditional vision models remain uninterpretable and memoryless, and while Vision-Language Models (VLMs) excel in semantic understanding, they rely on unsafe open-loop text reasoning lacking the anatomical grounding essential for clinical safety. Furthermore, robust learning is hindered by the scarcity of process supervision in sparse clinical records. To bridge this gap, we introduce the Logic-Constrained Abductive Data Engine. Operating on a ``Propose-and-Verify'' paradigm, it validates MLLM-proposed biomarkers against clinical and temporal logic to reconstruct dense pathological states from sparse outcomes. Building on this foundation, we propose ORBIT, the first ophthalmic Prognostic World Model. Uniquely, ORBIT employs counterfactual visual foresight to imagine anatomical futures under different treatments, anchoring decisions in Closed-Loop Anatomical Verification rather than linguistic probabilities. Experiments demonstrate that ORBIT effectively captures disease evolution and establishes a new paradigm for human-in-the-loop longitudinal decision support and anatomically grounded treatment planning.

Retinal diseases often require repeated visits and careful treatment decisions. A single eye scan may not be enough to tell whether a patient needs treatment or can safely continue observation. This paper presents ORBIT, an AI system that uses a patient’s past visits to understand disease changes over time and compare possible future outcomes under different treatment choices. ORBIT helps identify retinal disease signs, track progression, and support more balanced treatment recommendations. It is designed to assist, not replace, ophthalmologists, and still requires broader clinical validation before real-world use.