A Survey of Multimodal Ophthalmic Diagnostics: From Task-Specific Approaches to Foundational Models

TL;DR

This survey reviews recent advances in multimodal deep learning for ophthalmic diagnostics, highlighting task-specific methods and foundational models that improve disease detection, report generation, and decision support.

Contribution

It provides a comprehensive overview of the latest multimodal deep learning techniques, datasets, and challenges in ophthalmology up to 2025.

Findings

Task-specific approaches excel in clinical applications.

Foundation models enable cross-modal understanding and report automation.

Challenges include data variability and limited annotations.

Abstract

Visual impairment represents a major global health challenge, with multimodal imaging providing complementary information that is essential for accurate ophthalmic diagnosis. This comprehensive survey systematically reviews the latest advances in multimodal deep learning methods in ophthalmology up to the year 2025. The review focuses on two main categories: task-specific multimodal approaches and large-scale multimodal foundation models. Task-specific approaches are designed for particular clinical applications such as lesion detection, disease diagnosis, and image synthesis. These methods utilize a variety of imaging modalities including color fundus photography, optical coherence tomography, and angiography. On the other hand, foundation models combine sophisticated vision-language architectures and large language models pretrained on diverse ophthalmic datasets. These models enable robust cross-modal understanding, automated clinical report generation, and decision support. The survey critically examines important datasets, evaluation metrics, and methodological innovations including self-supervised learning, attention-based fusion, and contrastive alignment. It also discusses ongoing challenges such as variability in data, limited annotations, lack of interpretability, and issues with generalizability across different patient populations. Finally, the survey outlines promising future directions that emphasize the use of ultra-widefield imaging and reinforcement learning-based reasoning frameworks to create intelligent, interpretable, and clinically applicable AI systems for ophthalmology.