The impact of site-specific digital histology signatures on deep learning model accuracy and bias
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...
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article (Journal) |
| Language: | English |
| Published: |
20 July 2021
|
| In: |
Nature Communications
Year: 2021, Volume: 12, Pages: 1-13 |
| ISSN: | 2041-1723 |
| DOI: | 10.1038/s41467-021-24698-1 |
| Online Access: | Verlag, lizenzpflichtig, Volltext: https://doi.org/10.1038/s41467-021-24698-1 Verlag, lizenzpflichtig, Volltext: https://www.nature.com/articles/s41467-021-24698-1 
|
| Author Notes: | 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 & Alexander T. Pearson |
| Summary: | 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. |
|---|---|
| Item Description: | Gesehen am 07.10.2021 |
| Physical Description: | Online Resource |
| ISSN: | 2041-1723 |
| DOI: | 10.1038/s41467-021-24698-1 |