Medical Image Understanding with Pretrained Vision Language Models: A Comprehensive Study

TL;DR

This study explores how pre-trained vision language models can be effectively transferred to medical imaging, demonstrating that well-designed prompts and automatic prompt generation significantly enhance zero-shot and fine-tuned performance across diverse medical datasets.

Contribution

The paper introduces methods for automatic medical prompt generation and demonstrates their effectiveness in improving VLM performance in medical image understanding.

Findings

Well-designed prompts improve zero-shot accuracy.

Automatic prompt generation injects expert medical knowledge.

Fine-tuned models outperform supervised counterparts.

Abstract

The large-scale pre-trained vision language models (VLM) have shown remarkable domain transfer capability on natural images. However, it remains unknown whether this capability can also apply to the medical image domain. This paper thoroughly studies the knowledge transferability of pre-trained VLMs to the medical domain, where we show that well-designed medical prompts are the key to elicit knowledge from pre-trained VLMs. We demonstrate that by prompting with expressive attributes that are shared between domains, the VLM can carry the knowledge across domains and improve its generalization. This mechanism empowers VLMs to recognize novel objects with fewer or without image samples. Furthermore, to avoid the laborious manual designing process, we develop three approaches for automatic generation of medical prompts, which can inject expert-level medical knowledge and image-specific information into the prompts for fine-grained grounding. We conduct extensive experiments on thirteen different medical datasets across various modalities, showing that our well-designed prompts greatly improve the zero-shot performance compared to the default prompts, and our fine-tuned models surpass the supervised models by a significant margin.