MM-Retinal: Knowledge-Enhanced Foundational Pretraining with Fundus Image-Text Expertise
This work addresses the need for more transferable and generalizable models in medical imaging, specifically for fundus image analysis, by incorporating expert knowledge, though it appears incremental as it builds on existing foundational pretraining approaches.
The authors tackled the problem of poor transferability and generalizability in fundus image analysis by creating a multi-modal dataset and a knowledge-enhanced pretraining model, achieving state-of-the-art performance across six unseen downstream tasks with excellent generalization in zero-shot and few-shot scenarios.
Current fundus image analysis models are predominantly built for specific tasks relying on individual datasets. The learning process is usually based on data-driven paradigm without prior knowledge, resulting in poor transferability and generalizability. To address this issue, we propose MM-Retinal, a multi-modal dataset that encompasses high-quality image-text pairs collected from professional fundus diagram books. Moreover, enabled by MM-Retinal, we present a novel Knowledge-enhanced foundational pretraining model which incorporates Fundus Image-Text expertise, called KeepFIT. It is designed with image similarity-guided text revision and mixed training strategy to infuse expert knowledge. Our proposed fundus foundation model achieves state-of-the-art performance across six unseen downstream tasks and holds excellent generalization ability in zero-shot and few-shot scenarios. MM-Retinal and KeepFIT are available at https://github.com/lxirich/MM-Retinal.