Differentiable Multi-Agent Actor-Critic for Multi-Step Radiology Report Summarization

TL;DR

This paper introduces a two-step, explainable approach for radiology report summarization, involving extraction of salient sentences and keywords followed by abstractive summarization, improving precision over single-step models.

Contribution

It proposes a novel cascade model with separate extraction of sentences and keywords, enhancing interpretability and summarization accuracy in radiology reports.

Findings

Improved F1 score by 3-4% over baselines

Two-step extraction plus abstractive summarization outperforms single-step models

Enhanced explainability of the summarization process

Abstract

The IMPRESSIONS section of a radiology report about an imaging study is a summary of the radiologist's reasoning and conclusions, and it also aids the referring physician in confirming or excluding certain diagnoses. A cascade of tasks are required to automatically generate an abstractive summary of the typical information-rich radiology report. These tasks include acquisition of salient content from the report and generation of a concise, easily consumable IMPRESSIONS section. Prior research on radiology report summarization has focused on single-step end-to-end models -- which subsume the task of salient content acquisition. To fully explore the cascade structure and explainability of radiology report summarization, we introduce two innovations. First, we design a two-step approach: extractive summarization followed by abstractive summarization. Second, we additionally break down the extractive part into two independent tasks: extraction of salient (1) sentences and (2) keywords. Experiments on English radiology reports from two clinical sites show our novel approach leads to a more precise summary compared to single-step and to two-step-with-single-extractive-process baselines with an overall improvement in F1 score Of 3-4%.