The impact of site-specific digital histology signatures on deep learning model accuracy and bias

Frederick M. Howard, James Dolezal, Sara Kochanny, Jefree Schulte, Heather Chen, Lara Heij, Dezheng Huo, Rita Nanda, Olufunmilayo I. Olopade, Jakob N. Kather, Nicole Cipriani, Robert L. Grossman () and Alexander T. Pearson ()
Additional contact information
Frederick M. Howard: University of Chicago
James Dolezal: University of Chicago
Sara Kochanny: University of Chicago
Jefree Schulte: University of Chicago
Heather Chen: University of Chicago
Lara Heij: University Hospital RWTH Aachen
Dezheng Huo: University of Chicago
Rita Nanda: University of Chicago
Olufunmilayo I. Olopade: University of Chicago
Jakob N. Kather: University Hospital RWTH Aachen
Nicole Cipriani: University of Chicago
Robert L. Grossman: University of Chicago
Alexander T. Pearson: University of Chicago

Nature Communications, 2021, vol. 12, issue 1, 1-13

Abstract: Abstract The Cancer Genome Atlas (TCGA) is one of the largest biorepositories of digital histology. Deep learning (DL) models have been trained on TCGA to predict numerous features directly from histology, including survival, gene expression patterns, and driver mutations. However, we demonstrate that these features vary substantially across tissue submitting sites in TCGA for over 3,000 patients with six cancer subtypes. Additionally, we show that histologic image differences between submitting sites can easily be identified with DL. Site detection remains possible despite commonly used color normalization and augmentation methods, and we quantify the image characteristics constituting this site-specific digital histology signature. We demonstrate that these site-specific signatures lead to biased accuracy for prediction of features including survival, genomic mutations, and tumor stage. Furthermore, ethnicity can also be inferred from site-specific signatures, which must be accounted for to ensure equitable application of DL. These site-specific signatures can lead to overoptimistic estimates of model performance, and we propose a quadratic programming method that abrogates this bias by ensuring models are not trained and validated on samples from the same site.

Date: 2021
References: Add references at CitEc
Citations: View citations in EconPapers (3)

Downloads: (external link)
https://www.nature.com/articles/s41467-021-24698-1 Abstract (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nat:natcom:v:12:y:2021:i:1:d:10.1038_s41467-021-24698-1

Ordering information: This journal article can be ordered from
https://www.nature.com/ncomms/

DOI: 10.1038/s41467-021-24698-1

Access Statistics for this article

Nature Communications is currently edited by Nathalie Le Bot, Enda Bergin and Fiona Gillespie

More articles in Nature Communications from Nature
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().