Text-guided Foundation Model Adaptation for Long-Tailed Medical Image Classification

TL;DR

This paper introduces TFA-LT, a lightweight two-stage adaptation method that significantly improves long-tailed medical image classification by leveraging foundation models with minimal GPU resources.

Contribution

The paper proposes a novel, simple, and efficient two-stage adaptation strategy for foundation models tailored to long-tailed medical image classification tasks.

Findings

Achieves up to 27.1% accuracy improvement

Uses only 6.1% GPU memory of current best methods

Effective in handling data imbalance in medical datasets

Abstract

In medical contexts, the imbalanced data distribution in long-tailed datasets, due to scarce labels for rare diseases, greatly impairs the diagnostic accuracy of deep learning models. Recent multimodal text-image supervised foundation models offer new solutions to data scarcity through effective representation learning. However, their limited medical-specific pretraining hinders their performance in medical image classification relative to natural images. To address this issue, we propose a novel Text-guided Foundation model Adaptation for Long-Tailed medical image classification (TFA-LT). We adopt a two-stage training strategy, integrating representations from the foundation model using just two linear adapters and a single ensembler for balanced outcomes. Experimental results on two long-tailed medical image datasets validate the simplicity, lightweight and efficiency of our approach: requiring only 6.1% GPU memory usage of the current best-performing algorithm, our method achieves an accuracy improvement of up to 27.1%, highlighting the substantial potential of foundation model adaptation in this area.