CohortFinder: an open-source tool for data-driven partitioning of digital pathology and imaging cohorts to yield robust machine-learning models

Fan Fan; Georgia Martinez; Thomas DeSilvio; John Shin; Yijiang Chen; Jackson Jacobs; Bangchen Wang; Takaya Ozeki; Maxime W. Lafarge; Viktor H. Koelzer; Laura Barisoni; Anant Madabhushi; Satish E. Viswanath; Andrew Janowczyk

doi:10.1038/s44303-024-00018-2

npj Imaging (Jul 2024)

CohortFinder: an open-source tool for data-driven partitioning of digital pathology and imaging cohorts to yield robust machine-learning models

Fan Fan,
Georgia Martinez,
Thomas DeSilvio,
John Shin,
Yijiang Chen,
Jackson Jacobs,
Bangchen Wang,
Takaya Ozeki,
Maxime W. Lafarge,
Viktor H. Koelzer,
Laura Barisoni,
Anant Madabhushi,
Satish E. Viswanath,
Andrew Janowczyk

Affiliations

Fan Fan: Emory University and Georgia Institute of Technology, Department of Biomedical Engineering
Georgia Martinez: Case Western Reserve University, Department of Biomedical Engineering
Thomas DeSilvio: Case Western Reserve University, Department of Biomedical Engineering
John Shin: Case Western Reserve University, Department of Biomedical Engineering
Yijiang Chen: Case Western Reserve University, Department of Biomedical Engineering
Jackson Jacobs: Emory University and Georgia Institute of Technology, Department of Biomedical Engineering
Bangchen Wang: Duke University, Department of Pathology, Division of AI & Computational Pathology
Takaya Ozeki: University of Michigan, Department of Internal Medicine, Division of Nephrology
Maxime W. Lafarge: University Hospital of Zurich, University of Zurich, Department of Pathology and Molecular Pathology
Viktor H. Koelzer: University Hospital of Zurich, University of Zurich, Department of Pathology and Molecular Pathology
Laura Barisoni: Duke University, Department of Pathology, Division of AI & Computational Pathology
Anant Madabhushi: Emory University and Georgia Institute of Technology, Department of Biomedical Engineering
Satish E. Viswanath: Case Western Reserve University, Department of Biomedical Engineering
Andrew Janowczyk: Emory University and Georgia Institute of Technology, Department of Biomedical Engineering

DOI: https://doi.org/10.1038/s44303-024-00018-2
Journal volume & issue: Vol. 2, no. 1
pp. 1 – 7

Abstract

Read online

Abstract Batch effects (BEs) refer to systematic technical differences in data collection unrelated to biological variations whose noise is shown to negatively impact machine learning (ML) model generalizability. Here we release CohortFinder ( http://cohortfinder.com ), an open-source tool aimed at mitigating BEs via data-driven cohort partitioning. We demonstrate CohortFinder improves ML model performance in downstream digital pathology and medical image processing tasks. CohortFinder is freely available for download at cohortfinder.com.

Published in npj Imaging

ISSN: 2948-197X (Online)
Publisher: Nature Portfolio
Country of publisher: United Kingdom
LCC subjects: Medicine: Medicine (General): Medical technology; Medicine: Medicine (General): Medical physics. Medical radiology. Nuclear medicine
Website: https://www.nature.com/npjimaging/

About the journal