Multi-task UNet: Jointly Boosting Saliency Prediction and Disease Classification on Chest X-ray Images

TL;DR

This paper introduces a multi-task deep learning model that jointly predicts visual saliency and classifies diseases in chest X-ray images, improving performance through an optimized training scheme and addressing data scarcity.

Contribution

The paper presents a novel multi-task UNet model with an enhanced learning scheme for simultaneous saliency prediction and disease classification on chest X-rays.

Findings

Outperforms existing saliency prediction methods.

Achieves superior disease classification accuracy.

Demonstrates robustness with the new training scheme.

Abstract

Human visual attention has recently shown its distinct capability in boosting machine learning models. However, studies that aim to facilitate medical tasks with human visual attention are still scarce. To support the use of visual attention, this paper describes a novel deep learning model for visual saliency prediction on chest X-ray (CXR) images. To cope with data deficiency, we exploit the multi-task learning method and tackles disease classification on CXR simultaneously. For a more robust training process, we propose a further optimized multi-task learning scheme to better handle model overfitting. Experiments show our proposed deep learning model with our new learning scheme can outperform existing methods dedicated either for saliency prediction or image classification. The code used in this paper is available at https://github.com/hz-zhu/MT-UNet.