{"slug": "a-world-model-of-radiologist-reading-for-medical-image-representation-learning", "title": "A World Model of Radiologist Reading for Medical Image Representation Learning", "summary": "Researchers developed GazeWorld, a medical imaging model that learns from radiologists' eye-tracking data by treating their fixation sequences as trajectories through an image. The model autoregressively predicts latent representations of fixated patches and covers unvisited regions, achieving state-of-the-art diagnostic accuracy across nine supervised settings on three medical benchmarks. GazeWorld also outperformed purpose-built gaze prediction models by over 16% in ScanMatch and 22% in SED, demonstrating that modeling the process of expert image reading offers a promising pretraining paradigm for medical AI.", "body_md": "arXiv:2605.23992v1 Announce Type: new\nAbstract: Radiologist eye-tracking data provide a rich record of how experts search, compare, and accumulate evidence during image reading; yet, existing methods exploit this signal only partially, either as a static spatial prior or as an auxiliary prediction target decoupled from diagnosis. We propose GazeWorld, a medical imaging world model that treats the image as the world and the radiologist's fixation sequence as a trajectory through it. GazeWorld autoregressively predicts the latent representation of the next fixated patch from all previously visited ones, while a spatial-completion branch covers unvisited regions. At inference, GazeWorld generates a sequence of patch representations from the image alone without requiring real gaze data. Frozen GazeWorld features achieve state-of-the-art diagnostic accuracy across all nine supervised settings on CheXpert, RSNA Pneumonia, and SIIM-ACR Pneumothorax, as well as the highest zero-shot accuracy on all three benchmarks. On the GazeSearch benchmark, a generic decoder trained on the same frozen features outperforms the purpose-built LogitGaze-Med by over 16\\% in ScanMatch and 22\\% in SED, despite not being explicitly trained to predict gaze. GazeWorld demonstrates that modeling how experts read, not just what they conclude, offers a promising pretraining paradigm for medical imaging AI.", "url": "https://wpnews.pro/news/a-world-model-of-radiologist-reading-for-medical-image-representation-learning", "canonical_source": "https://arxiv.org/abs/2605.23992", "published_at": "2026-05-26 04:00:00+00:00", "updated_at": "2026-05-26 04:10:46.677785+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "computer-vision", "ai-research", "neural-networks"], "entities": ["GazeWorld", "CheXpert", "RSNA Pneumonia", "SIIM-ACR Pneumothorax", "GazeSearch", "LogitGaze-Med"], "alternates": {"html": "https://wpnews.pro/news/a-world-model-of-radiologist-reading-for-medical-image-representation-learning", "markdown": "https://wpnews.pro/news/a-world-model-of-radiologist-reading-for-medical-image-representation-learning.md", "text": "https://wpnews.pro/news/a-world-model-of-radiologist-reading-for-medical-image-representation-learning.txt", "jsonld": "https://wpnews.pro/news/a-world-model-of-radiologist-reading-for-medical-image-representation-learning.jsonld"}}