CatScreen: A Large MultiModal Benchmark Dataset for Cataract Screening

TMLR Paper5513 Authors

Abstract: Low-cost slit-lamp imaging holds significant potential for transforming eye care by facilitating affordable and scalable cataract diagnosis. However, the development of robust, generalizable AI-based cataract screening solutions is currently constrained by the limited availability of large-scale, richly annotated datasets. To address this critical gap, we introduce CatScreen, a comprehensive multimodal benchmark dataset specifically designed for cataract screening, comprising approximately 18,000 slit-lamp images collected from 2,251 subjects using a portable slit-lamp camera. CatScreen is structured into three subsets: (i) a clean set meticulously annotated by ophthalmology experts across clinically relevant dimensions, including image gradability, quality assessment, illumination type, diagnostic classification, cataract subtype, and severity grading according to established standards; (ii) a noisy-labeled set that simulates real-world annotation inaccuracies; and (iii) an unlabeled set intended to foster the development of self-supervised and semi-supervised learning approaches. Furthermore, CatScreen integrates extensive subject-level metadata encompassing demographics, lifestyle factors, and detailed clinical histories, providing a holistic perspective for comprehensive analysis. To enhance model interpretability and clinical applicability, a subset of images has been precisely annotated to delineate anatomical structures in both healthy and pathological states. Additionally, this work presents two complementary AI frameworks, Structured Sequential Analysis and Multitask Learning, each offering distinct yet synergistic approaches toward enhancing model interpretability and efficiency. CatScreen thus provides researchers with a robust foundation to advance reliable, interpretable, and generalizable cataract screening solutions, significantly improving access to quality eye care diagnostics, particularly in underserved and resource-limited regions.

Submission Length: Long submission (more than 12 pages of main content)

Assigned Action Editor: ~ERIC_EATON1

Submission Number: 5513