Deceptive learning in histopathology
Histopathology · 2024-07 · 85 (1) · 116-132
DOI: 10.1111/his.15180 ↗Kurz gefasst
Fachartikel in der Zeitschrift Histopathology, online erschienen am 31.03.2024. Verfasst von Sahar Shahamatdar und 8 weiteren Autor:innen. Beteiligte Einrichtung: Center for Computational Molecular Biology, Brown University, Providence, RI, USA. Hauptthemen laut MEDLINE: Deep Learning, Lung Neoplasms, Adenocarcinoma of Lung. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIH HHS, Moffitt Cancer Center, Brown University.
Abstract (Originalsprache)
Aims
Deep learning holds immense potential for histopathology, automating tasks that are simple for expert pathologists and revealing novel biology for tasks that were previously considered difficult or impossible to solve by eye alone. However, the extent to which the visual strategies learned by deep learning models in histopathological analysis are trustworthy or not has yet to be systematically analysed. Here, we systematically evaluate deep neural networks (DNNs) trained for histopathological analysis in order to understand if their learned strategies are trustworthy or deceptive. Methods and results We trained a variety of DNNs on a novel data set of 221 whole‐slide images (WSIs) from lung adenocarcinoma patients, and evaluated their effectiveness at (1) molecular profiling of KRAS versus EGFR mutations, (2) determining the primary tissue of a tumour and (3) tumour detection. While DNNs achieved above‐chance performance on molecular profiling, they did so by exploiting correlations between histological subtypes and mutations, and failed to generalise to a challenging test set obtained through laser capture microdissection (LCM). In contrast, DNNs learned robust and trustworthy strategies for determining the primary tissue of a tumour as well as detecting and localising tumours in tissue.
Conclusions
Our work demonstrates that DNNs hold immense promise for aiding pathologists in analysing tissue. However, they are also capable of achieving seemingly strong performance by learning deceptive strategies that leverage spurious correlations, and are ultimately unsuitable for research or clinical work. The framework we propose for model evaluation and interpretation is an important step towards developing reliable automated systems for histopathological analysis.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Center for Computational Molecular Biology, Brown University, Providence, RI, USA.
- The Warren Alpert Medical School, Brown University, Providence, RI, USA.
- Department of Anatomic Pathology, H. Lee Moffitt Cancer and Research Institute, Tampa, FL, USA.
- Carney Institute for Brain Science, Brown University, Providence, RI, USA.
- Department of Cognitive Linguistic and Psychological Sciences, Brown University, Providence, RI, USA.
- Department of Molecular Biology, H. Lee Moffitt Cancer and Research Institute, Tampa, FL, USA.
ORCID-Profile: Daryoush Saeed-Vafa
Förderung
- NIH HHS
- Moffitt Cancer Center
- Brown University
- NIGMS NIH HHS
- National Institute of General Medical Sciences
- National Institutes of Health
Zitieren
APA
Shahamatdar, S., Saeed-Vafa, D., Linsley, D., Khalil, F., Lovinger, K., Li, L., McLeod, H. T., Ramachandran, S., & Serre, T. (2024). Deceptive learning in histopathology. Histopathology, 85(1), 116-132. https://doi.org/10.1111/his.15180
Vancouver
Shahamatdar S, Saeed-Vafa D, Linsley D, Khalil F, Lovinger K, Li L, et al. Deceptive learning in histopathology. Histopathology. 2024;85(1):116-132. doi:10.1111/his.15180
BibTeX
@article{Shahamatdar2024,
author = {Shahamatdar, Sahar and Saeed-Vafa, Daryoush and Linsley, Drew and Khalil, Farah and Lovinger, Katherine and Li, Lester and McLeod, Howard T and Ramachandran, Sohini and Serre, Thomas},
title = {Deceptive learning in histopathology},
journal = {Histopathology},
year = {2024},
volume = {85},
number = {1},
pages = {116--132},
doi = {10.1111/his.15180}
}
EUVIMED