Weakly Supervised Contrastive Learning for Chest X-Ray Report Generation

TL;DR

This paper introduces a weakly supervised contrastive learning approach for chest X-ray report generation, improving clinical accuracy and report quality by contrasting correct reports with semantically similar incorrect ones.

Contribution

It proposes a novel contrastive loss function tailored for medical report generation, enhancing the quality of generated reports over traditional methods.

Findings

Outperforms previous methods on clinical correctness metrics

Improves text generation quality on public benchmarks

Utilizes contrasting with semantically-close incorrect reports

Abstract

Radiology report generation aims at generating descriptive text from radiology images automatically, which may present an opportunity to improve radiology reporting and interpretation. A typical setting consists of training encoder-decoder models on image-report pairs with a cross entropy loss, which struggles to generate informative sentences for clinical diagnoses since normal findings dominate the datasets. To tackle this challenge and encourage more clinically-accurate text outputs, we propose a novel weakly supervised contrastive loss for medical report generation. Experimental results demonstrate that our method benefits from contrasting target reports with incorrect but semantically-close ones. It outperforms previous work on both clinical correctness and text generation metrics for two public benchmarks.