A Large-scale Medical Visual Task Adaptation Benchmark

TL;DR

This paper introduces Med-VTAB, a large-scale benchmark for medical visual task adaptation, and proposes GMoE-Adapter, a novel method that improves adaptation performance across diverse medical imaging modalities.

Contribution

It provides the first large-scale benchmark for medical visual task adaptation and introduces GMoE-Adapter, a new approach combining pre-trained weights for better performance.

Findings

Single pre-trained models are insufficient for medical task adaptation.

GMoE-Adapter achieves state-of-the-art results in medical visual adaptation.

Scaling laws and out-of-distribution effects are analyzed.

Abstract

Visual task adaptation has been demonstrated to be effective in adapting pre-trained Vision Transformers (ViTs) to general downstream visual tasks using specialized learnable layers or tokens. However, there is yet a large-scale benchmark to fully explore the effect of visual task adaptation on the realistic and important medical domain, particularly across diverse medical visual modalities, such as color images, X-ray, and CT. To close this gap, we present Med-VTAB, a large-scale Medical Visual Task Adaptation Benchmark consisting of 1.68 million medical images for diverse organs, modalities, and adaptation approaches. Based on Med-VTAB, we explore the scaling law of medical prompt tuning concerning tunable parameters and the generalizability of medical visual adaptation using non-medical/medical pre-train weights. Besides, we study the impact of patient ID out-of-distribution on medical visual adaptation, which is a real and challenging scenario. Furthermore, results from Med-VTAB indicate that a single pre-trained model falls short in medical task adaptation. Therefore, we introduce GMoE-Adapter, a novel method that combines medical and general pre-training weights through a gated mixture-of-experts adapter, achieving state-of-the-art results in medical visual task adaptation.