MLIP: Enhancing Medical Visual Representation with Divergence Encoder and Knowledge-guided Contrastive Learning

TL;DR

MLIP introduces a domain-specific contrastive learning framework that integrates medical knowledge and multi-granularity image-text alignment to improve medical visual representations, especially with limited annotated data.

Contribution

It proposes a novel divergence encoder and knowledge-guided contrastive learning techniques to better utilize image-text information across multiple granularities in medical imaging.

Findings

Outperforms state-of-the-art methods in medical image classification, detection, and segmentation.

Enhances transfer performance with limited annotated data.

Demonstrates the effectiveness of multimodal pre-training in medical imaging.

Abstract

The scarcity of annotated data has sparked significant interest in unsupervised pre-training methods that leverage medical reports as auxiliary signals for medical visual representation learning. However, existing research overlooks the multi-granularity nature of medical visual representation and lacks suitable contrastive learning techniques to improve the models' generalizability across different granularities, leading to the underutilization of image-text information. To address this, we propose MLIP, a novel framework leveraging domain-specific medical knowledge as guiding signals to integrate language information into the visual domain through image-text contrastive learning. Our model includes global contrastive learning with our designed divergence encoder, local token-knowledge-patch alignment contrastive learning, and knowledge-guided category-level contrastive learning with expert knowledge. Experimental evaluations reveal the efficacy of our model in enhancing transfer performance for tasks such as image classification, object detection, and semantic segmentation. Notably, MLIP surpasses state-of-the-art methods even with limited annotated data, highlighting the potential of multimodal pre-training in advancing medical representation learning.