Generative Residual Attention Network for Disease Detection

TL;DR

This paper introduces a novel generative residual attention network that enhances disease detection in radiology images by generating and localizing abnormalities, improving detection accuracy with limited annotated data.

Contribution

It presents a combined framework for disease generation and localization using conditional GANs, advancing medical image analysis with limited annotations.

Findings

Outperforms state-of-the-art detection algorithms on RSNA dataset

Effective data augmentation improves disease detection accuracy

Simultaneous disease generation and localization enhances model robustness

Abstract

Accurate identification and localization of abnormalities from radiology images serve as a critical role in computer-aided diagnosis (CAD) systems. Building a highly generalizable system usually requires a large amount of data with high-quality annotations, including disease-specific global and localization information. However, in medical images, only a limited number of high-quality images and annotations are available due to annotation expenses. In this paper, we explore this problem by presenting a novel approach for disease generation in X-rays using a conditional generative adversarial learning. Specifically, given a chest X-ray image from a source domain, we generate a corresponding radiology image in a target domain while preserving the identity of the patient. We then use the generated X-ray image in the target domain to augment our training to improve the detection performance. We also present a unified framework that simultaneously performs disease generation and localization.We evaluate the proposed approach on the X-ray image dataset provided by the Radiological Society of North America (RSNA), surpassing the state-of-the-art baseline detection algorithms.