Segmentation of Lungs in Chest X-Ray Image Using Generative Adversarial Networks

Faizan Munawar,Hazrat Ali,Shoaib Azmat,Talha Iqbal,Christer Gronlund

doi:10.1109/access.2020.3017915

Abstract

Chest X-ray (CXR) is a low-cost medical imaging technique. It is a common procedure for the identification of many respiratory diseases compared to MRI, CT, and PET scans. This paper presents the use of generative adversarial networks (GAN) to perform the task of lung segmentation on a given CXR. GANs are popular to generate realistic data by learning the mapping from one domain to another. In our work, the generator of the GAN is trained to generate a segmented mask of a given input CXR. The discriminator distinguishes between a ground truth and the generated mask, and updates the generator through the adversarial loss measure. The objective is to generate masks for the input CXR, which are as realistic as possible compared to the ground truth masks. The model is trained and evaluated using four different discriminators referred to as D1, D2, D3, and D4, respectively. Experimental results on three different CXR datasets reveal that the proposed model is able to achieve a dice-score of 0.9740, and IOU score of 0.943, which are better than other reported state-of-the art results.

Highlights

With ever increasing capabilities of modern technologies, the role of computer-aided diagnosis (CAD) systems has reached to a maximum significance than ever before [50]
CAD systems are used in various areas of medical diagnosis such as in mammography, for detection of breast cancer, polyps detection in colon, diabetic retinopathy, The associate editor coordinating the review of this manuscript and approving it for publication was Gang Mei
For Japanese Society of Radiological Technology’’ (JSRT) dataset, 200 images are used for training, 20 are used for cross validation, and 20 images were used for testing

Summary

Introduction

With ever increasing capabilities of modern technologies, the role of computer-aided diagnosis (CAD) systems has reached to a maximum significance than ever before [50]. CAD systems provide aid to physicians and health care professionals to better understand the clinical conditions of a patient and help in diagnosis of various diseases. Medical image procedures such as magnetic resonance imaging (MRI), X-rays and ultrasound imaging carry useful information, but require thorough study by an expert. CAD systems are used in various areas of medical diagnosis such as in mammography, for detection of breast cancer, polyps detection in colon, diabetic retinopathy, The associate editor coordinating the review of this manuscript and approving it for publication was Gang Mei

Methods

Results

Conclusion