Multi-task Learning for Chest X-ray Abnormality Classification on Noisy Labels

TL;DR

This paper introduces a multi-task deep learning system for chest X-ray abnormality classification that leverages concurrent segmentation and localization tasks to improve accuracy, achieving state-of-the-art results despite noisy labels.

Contribution

The work presents a novel multi-task architecture that enhances classification performance by jointly training on segmentation and localization tasks using a large noisy dataset.

Findings

Achieved 0.883 AUC on 12 abnormalities with extensive data.

Outperformed radiologists on a high-confidence subset.

Reaching 0.945 AUC with high-quality labels.

Abstract

Chest X-ray (CXR) is the most common X-ray examination performed in daily clinical practice for the diagnosis of various heart and lung abnormalities. The large amount of data to be read and reported, with 100+ studies per day for a single radiologist, poses a challenge in maintaining consistently high interpretation accuracy. In this work, we propose a method for the classification of different abnormalities based on CXR scans of the human body. The system is based on a novel multi-task deep learning architecture that in addition to the abnormality classification, supports the segmentation of the lungs and heart and classification of regions where the abnormality is located. We demonstrate that by training these tasks concurrently, one can increase the classification performance of the model. Experiments were performed on an extensive collection of 297,541 chest X-ray images from 86,876 patients, leading to a state-of-the-art performance level of 0.883 AUC on average for 12 different abnormalities. We also conducted a detailed performance analysis and compared the accuracy of our system with 3 board-certified radiologists. In this context, we highlight the high level of label noise inherent to this problem. On a reduced subset containing only cases with high confidence reference labels based on the consensus of the 3 radiologists, our system reached an average AUC of 0.945.