A team of researchers have developed a method of generating high-quality chest X-ray images that can be used to diagnose COVID-19 more accurately than current methods. The team, led by Sumeet Menon, a PhD student in computer science at UMBC, includes IDSC Chief Innovation Officer and Head of International Relations, Dr. Yelena Yesha, as well as IDSC Affiliate Member Dr. Michael Morris. Congratulations to the team for the publications of its findings in the highly prestigious Proceedings of the IEEE International Conference on Big Data (aka IEEE Big Data 2020) held virtually in December 2020.
COVID-19 is a novel infectious disease responsible for over 800K deaths worldwide as of August 2020. The need for rapid testing is a high priority, and alternative testing strategies including X-ray image classification are a promising area of research. However, at present, public datasets for COVID19 x-ray images have low data volumes, making it challenging to develop accurate image classifiers.
Several recent papers have made use of Generative Adversarial Networks (GANs) in order to increase the training data volumes. But realistic synthetic COVID19 X-rays remain challenging to generate. We present a novel Mean Teacher + Transfer GAN (MTT-GAN) that generates COVID19 chest X-ray images of high quality. In order to create a more accurate GAN, we employ transfer learning from the Kaggle Pneumonia X-Ray dataset, a highly relevant data source orders of magnitude larger than public COVID19 datasets. Furthermore, we employ the Mean Teacher algorithm as a constraint to improve stability of training.
Our qualitative analysis shows that the MTT-GAN generates X-ray images that are greatly superior to a baseline GAN and visually comparable to real X-rays. Although board-certified radiologists can distinguish MTT-GAN fakes from real COVID19 X-rays. Quantitative analysis shows that MTT-GAN greatly improves the accuracy of both a binary COVID19 classifier as well as a multi-class Pneumonia classifier as compared to a baseline GAN. Our classification accuracy is favorable as compared to recently reported results in the literature for similar binary and multi-class COVID19 screening tasks.
S. Menon, J. Galita, D. Chapman, A. Gangopadhyay, J. Mangalagiri, P. Nguyen, Y. Yesha, Y. Yesha, B. Saboury, M. Morris., “Generating Realistic COVID19 X-rays with a Mean Teacher + Transfer Learning GAN,” 2020, https://arxiv.org/pdf/2009.12478.pdf
“UMBC researchers use machine learning to develop more accurate COVID-19 diagnostic tool,” University of Maryland Baltimore County News Release December 10, 2020