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

dc.contributor.author

Cooke, Colin L

dc.contributor.author

Kim, Kanghyun

dc.contributor.author

Xu, Shiqi

dc.contributor.author

Chaware, Amey

dc.contributor.author

Yao, Xing

dc.contributor.author

Yang, Xi

dc.contributor.author

Neff, Jadee

dc.contributor.author

Pittman, Patricia

dc.contributor.author

McCall, Chad

dc.contributor.author

Glass, Carolyn

dc.contributor.author

Jiang, Xiaoyin Sara

dc.contributor.author

Horstmeyer, Roarke

dc.contributor.editor

Yoon, Dukyong

dc.date.accessioned

2024-05-14T19:06:14Z

dc.date.available

2024-05-14T19:06:14Z

dc.date.issued

2022-08

dc.description.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.

dc.identifier

PDIG-D-22-00017

dc.identifier.issn

2767-3170

dc.identifier.issn

2767-3170

dc.identifier.uri

https://hdl.handle.net/10161/30704

dc.language

eng

dc.publisher

Public Library of Science (PLoS)

dc.relation.ispartof

PLOS digital health

dc.relation.isversionof

10.1371/journal.pdig.0000078

dc.rights.uri

https://creativecommons.org/licenses/by-nc/4.0

dc.title

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

dc.type

Journal article

duke.contributor.orcid

Neff, Jadee|0000-0002-4924-4247

duke.contributor.orcid

Glass, Carolyn|0000-0002-8850-9906

duke.contributor.orcid

Horstmeyer, Roarke|0000-0002-2480-9141

pubs.begin-page

e0000078

pubs.issue

8

pubs.organisational-group

Duke

pubs.organisational-group

Pratt School of Engineering

pubs.organisational-group

School of Medicine

pubs.organisational-group

Clinical Science Departments

pubs.organisational-group

Institutes and Centers

pubs.organisational-group

Biomedical Engineering

pubs.organisational-group

Pathology

pubs.organisational-group

Surgery

pubs.organisational-group

Surgery, Cardiovascular and Thoracic Surgery

pubs.organisational-group

Duke Cancer Institute

pubs.organisational-group

University Institutes and Centers

pubs.organisational-group

Duke Institute for Brain Sciences

pubs.publication-status

Published

pubs.volume

1

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
A multiple instance learning approach for detecting COVID-19 in peripheral blood smears.pdf
Size:
2.64 MB
Format:
Adobe Portable Document Format