BERTHop: An Effective Vision-and-Language Model for Chest X-ray Disease Diagnosis

TL;DR

BERTHop is a transformer-based vision-and-language model tailored for chest X-ray disease diagnosis, effectively capturing modality associations and outperforming state-of-the-art models on a standard benchmark.

Contribution

The paper introduces BERTHop, a novel model combining PixelHop++ and VisualBERT, specifically designed to improve medical V&L tasks by addressing domain-specific challenges.

Findings

BERTHop achieves 98.12% AUC on OpenI dataset.

It outperforms SOTA by 1.62% AUC.

Requires 9 times less training data.

Abstract

Vision-and-language(V&L) models take image and text as input and learn to capture the associations between them. Prior studies show that pre-trained V&L models can significantly improve the model performance for downstream tasks such as Visual Question Answering (VQA). However, V&L models are less effective when applied in the medical domain (e.g., on X-ray images and clinical notes) due to the domain gap. In this paper, we investigate the challenges of applying pre-trained V&L models in medical applications. In particular, we identify that the visual representation in general V&L models is not suitable for processing medical data. To overcome this limitation, we propose BERTHop, a transformer-based model based on PixelHop++ and VisualBERT, for better capturing the associations between the two modalities. Experiments on the OpenI dataset, a commonly used thoracic disease diagnosis benchmark, show that BERTHop achieves an average Area Under the Curve (AUC) of 98.12% which is 1.62% higher than state-of-the-art (SOTA) while it is trained on a 9 times smaller dataset.