Semi-Supervised Classification of Noisy, Gigapixel Histology Images.

Journal: Proceedings. IEEE International Symposium on Bioinformatics and Bioengineering
Published Date: Dec 16, 2020

Abstract

One of the greatest obstacles in the adoption of deep neural networks for new medical applications is that training these models typically require a large amount of manually labeled training samples. In this body of work, we investigate the semi-supervised scenario where one has access to large amounts of unlabeled data and only a few labeled samples. We study the performance of MixMatch and FixMatch-two popular semi-supervised learning methods-on a histology dataset. More specifically, we study these models' impact under a highly noisy and imbalanced setting. The findings here motivate the development of semi-supervised methods to ameliorate problems commonly encountered in medical data applications.

Authors

  • J Vince Pulido
    Applied Physics Laboratory, Johns Hopkins University, Laurel, MD.
  • Shan Guleria
    Dept. of Internal Medicine, Rush University Medical Center, Chicago, IL.
  • Lubaina Ehsan
    School of Medicine, University of Virginia, Charlottesville, VA.
  • Matthew Fasullo
    Division of Gastroenterology, Hepatology and Nutrition, Virginia Commonwealth University, Richmond, VA.
  • Robert Lippman
    Hunter Holmes McGuire, Veterans Affairs Medical Center, Richmond, VA.
  • Pritesh Mutha
    Hunter Holmes McGuire, Veterans Affairs Medical Center, Richmond, VA.
  • Tilak Shah
    Hunter Holmes McGuire, Veterans Affairs Medical Center, Richmond, VA.
  • Sana Syed
    School of Medicine, University of Virginia, Charlottesville, VA.
  • Donald E Brown
    School of Data Science, University of Virginia, Charlottesville, VA.

Keywords

No keywords available for this article.

External Resources

Popular Topics
Recent Journals