A multiple instance learning approach for detecting COVID-19 in peripheral blood smears.

Abstract

A wide variety of diseases are commonly diagnosed via the visual examination of cell morphology within a peripheral blood smear. For certain diseases, such as COVID-19, morphological impact across the multitude of blood cell types is still poorly understood. In this paper, we present a multiple instance learning-based approach to aggregate high-resolution morphological information across many blood cells and cell types to automatically diagnose disease at a per-patient level. We integrated image and diagnostic information from across 236 patients to demonstrate not only that there is a significant link between blood and a patient's COVID-19 infection status, but also that novel machine learning approaches offer a powerful and scalable means to analyze peripheral blood smears. Our results both backup and enhance hematological findings relating blood cell morphology to COVID-19, and offer a high diagnostic efficacy; with a 79% accuracy and a ROC-AUC of 0.90.

Department

Description

Provenance

Subjects

Citation

Published Version (Please cite this version)

10.1371/journal.pdig.0000078

Publication Info

Cooke, Colin L, Kanghyun Kim, Shiqi Xu, Amey Chaware, Xing Yao, Xi Yang, Jadee Neff, Patricia Pittman, et al. (2022). A multiple instance learning approach for detecting COVID-19 in peripheral blood smears. PLOS digital health, 1(8). p. e0000078. 10.1371/journal.pdig.0000078 Retrieved from https://hdl.handle.net/10161/30704.

This is constructed from limited available data and may be imprecise. To cite this article, please review & use the official citation provided by the journal.

Scholars@Duke

Neff

Jadee Lee Neff

Assistant Professor of Pathology

As a diagnostic hematopathologist and molecular genetic pathologist, my clinical interests are focused on the histologic examination of tissue and bone marrow biopsies to diagnose hematologic malignancies (leukemia, lymphoma, myeloma, etc.) as well as testing DNA from tumors or from blood to detect inherited or acquired mutations that can guide therapeutic management and predict clinical outcome.  My research interests involve 1) understanding the biology of T-cell and NK-cell neoplasms; 2) defining the immunomodulatory response to neoplastic disease; 3) developing methods to monitor immune response and thereby refine tumor immunotherapy; and 4) exploring novel applications of tumor genetics in the diagnosis, prognosis, and management of cancer.

Glass

Carolyn Glass

Associate Professor of Pathology

Cardiothoracic Pathologist and Physician-Scientist
Division Chief, Cardiovascular Pathology 
Co-Director, Division of Artificial Intelligence and Computational Pathology
Director, Duke University Hospital Autopsy Service 
Associate Director, Residency Program  

Dr. Glass completed medical residency in Anatomic Pathology at the Brigham and Women’s Hospital/Harvard Medical School followed by fellowships in Cardiothoracic Pathology also at Brigham and Women’s Hospital/Harvard Medical School and Pulmonary/Cardiac Transplant Pathology at the University of Texas Southwestern Medical Center. Dr. Glass initially trained as a vascular surgeon with a focus on endovascular/interventional procedures through the 0+5 Integrated Vascular Surgery Program at the University of Rochester Medical Center from 2007-2011.  As a recipient of the NIH National Lung Blood Institute T32 Ruth Kirschstein National Service Research Award, she completed a Ph.D with a concentration in Genomics and Epigenetics in 2014.

Dr. Glass was awarded a five-year $3.2 million NIH grant to serve as P.I. of the Pathology Core as part of a larger U54 NIH grant ($13.5 million along with Duke Department of Medicine) to establish a Senescent Cell Human Tissue Mapping Center as part of the NIH Cellular Senescence Network. As a thoracic pathologist, Dr. Glass also has a special interest in identifying new epigenetic biomarkers that may predict response or resistance to conventional, targeted and immune therapy using computational techniques. She works closely with the Duke Thoracic Oncology Group, DCI Center for Cancer Immunotherapy, Duke Division of Cardiovascular Medicine and Cardiothoracic Surgery and Pratt School of Biomedical Engineering. 

Dr. Glass is the recipient of the Society of Cardiovascular Pathology (SCVP) Young Investigator’s Award, the William von Liebig Vascular Biology Research Fellowship at the Harvard Institutes of Medicine, the Duke Pathology Salvatore V. Pizzo Faculty Research Mentor Award, the Duke Department of Pathology Early Career Research Achievement Award and is author of over 90 publications (including book chapters in the recent W.H.O. Classification Tumours of the Lung, Pleura, Thymus and Heart) and 50 national presentations in cardiovascular disease, thoracic malignancies, surgery and machine learning. 

In addition to her clinical and research activities, Dr. Glass serves on the Executive/National Committees for the Society of Cardiovascular Pathology, College of American Pathology Artificial Intelligence Committee and the Duke School of Medicine Executive Admissions Committee. 




Horstmeyer

Roarke Horstmeyer

Assistant Professor of Biomedical Engineering

Roarke Horstmeyer is an assistant professor within Duke's Biomedical Engineering Department. He develops microscopes, cameras and computer algorithms for a wide range of applications, from forming 3D reconstructions of organisms to detecting neural activity deep within tissue. His areas of interest include optics, signal processing, optimization and neuroscience. Most recently, Dr. Horstmeyer was a guest professor at the University of Erlangen in Germany and an Einstein postdoctoral fellow at Charitè Medical School in Berlin. Prior to his time in Germany, Dr. Horstmeyer earned a PhD from Caltech’s electrical engineering department in 2016, a master of science degree from the MIT Media Lab in 2011, and a bachelors degree in physics and Japanese from Duke University in 2006.


Unless otherwise indicated, scholarly articles published by Duke faculty members are made available here with a CC-BY-NC (Creative Commons Attribution Non-Commercial) license, as enabled by the Duke Open Access Policy. If you wish to use the materials in ways not already permitted under CC-BY-NC, please consult the copyright owner. Other materials are made available here through the author’s grant of a non-exclusive license to make their work openly accessible.