A Dual-View Approach to Classifying Radiology Reports by Co-Training

TL;DR

This paper introduces a dual-view co-training method that leverages the distinct structural sections of radiology reports to improve classification accuracy using semi-supervised learning, outperforming existing methods.

Contribution

The study presents a novel co-training approach utilizing the Findings and Impression sections as separate views for semi-supervised radiology report classification.

Findings

Improved classification performance over baseline methods

Effective use of unlabeled data in semi-supervised setting

Outperforms existing supervised and semi-supervised approaches

Abstract

Radiology report analysis provides valuable information that can aid with public health initiatives, and has been attracting increasing attention from the research community. In this work, we present a novel insight that the structure of a radiology report (namely, the Findings and Impression sections) offers different views of a radiology scan. Based on this intuition, we further propose a co-training approach, where two machine learning models are built upon the Findings and Impression sections, respectively, and use each other's information to boost performance with massive unlabeled data in a semi-supervised manner. We conducted experiments in a public health surveillance study, and results show that our co-training approach is able to improve performance using the dual views and surpass competing supervised and semi-supervised methods.