Clinically Accurate Chest X-Ray Report Generation

TL;DR

This paper introduces a domain-aware system for generating clinically accurate chest X-ray reports, combining topic prediction and conditional sentence generation, fine-tuned with reinforcement learning to enhance clinical relevance and readability.

Contribution

The work presents a novel approach that integrates topic prediction with conditional report generation and employs reinforcement learning for clinical accuracy, addressing limitations of prior methods.

Findings

Improved language generation metrics over baselines

Enhanced clinical accuracy as per CheXpert scores

Validated on Open-I and MIMIC-CXR datasets

Abstract

The automatic generation of radiology reports given medical radiographs has significant potential to operationally and improve clinical patient care. A number of prior works have focused on this problem, employing advanced methods from computer vision and natural language generation to produce readable reports. However, these works often fail to account for the particular nuances of the radiology domain, and, in particular, the critical importance of clinical accuracy in the resulting generated reports. In this work, we present a domain-aware automatic chest X-ray radiology report generation system which first predicts what topics will be discussed in the report, then conditionally generates sentences corresponding to these topics. The resulting system is fine-tuned using reinforcement learning, considering both readability and clinical accuracy, as assessed by the proposed Clinically Coherent Reward. We verify this system on two datasets, Open-I and MIMIC-CXR, and demonstrate that our model offers marked improvements on both language generation metrics and CheXpert assessed accuracy over a variety of competitive baselines.