Interpreting chest X-rays via CNNs that exploit hierarchical disease dependencies and uncertainty labels

TL;DR

This paper introduces a deep CNN framework for multi-label classification of 14 thoracic diseases in chest X-rays, leveraging hierarchical label dependencies and uncertainty handling, achieving state-of-the-art results on the CheXpert dataset.

Contribution

It proposes a novel CNN-based method that exploits disease label dependencies and uses label smoothing to handle uncertain samples, improving multi-disease detection accuracy.

Findings

Achieved a mean AUC of 0.940 on the validation set.

Outperformed 2.6 out of 3 radiologists on the test set.

Ranked first on the CheXpert leaderboard at the time of publication.

Abstract

Chest radiography is one of the most common types of diagnostic radiology exams, which is critical for screening and diagnosis of many different thoracic diseases. Specialized algorithms have been developed to detect several specific pathologies such as lung nodule or lung cancer. However, accurately detecting the presence of multiple diseases from chest X-rays (CXRs) is still a challenging task. This paper presents a supervised multi-label classification framework based on deep convolutional neural networks (CNNs) for predicting the risk of 14 common thoracic diseases. We tackle this problem by training state-of-the-art CNNs that exploit dependencies among abnormality labels. We also propose to use the label smoothing technique for a better handling of uncertain samples, which occupy a significant portion of almost every CXR dataset. Our model is trained on over 200,000 CXRs of the recently released CheXpert dataset and achieves a mean area under the curve (AUC) of 0.940 in predicting 5 selected pathologies from the validation set. This is the highest AUC score yet reported to date. The proposed method is also evaluated on the independent test set of the CheXpert competition, which is composed of 500 CXR studies annotated by a panel of 5 experienced radiologists. The performance is on average better than 2.6 out of 3 other individual radiologists with a mean AUC of 0.930, which ranks first on the CheXpert leaderboard at the time of writing this paper.