Repository logo
 

Enabling large-scale screening of Barrett's esophagus using weakly supervised deep learning in histopathology.

Published version
Peer-reviewed

Repository DOI


Change log

Abstract

Timely detection of Barrett's esophagus, the pre-malignant condition of esophageal adenocarcinoma, can improve patient survival rates. The Cytosponge-TFF3 test, a non-endoscopic minimally invasive procedure, has been used for diagnosing intestinal metaplasia in Barrett's. However, it depends on pathologist's assessment of two slides stained with H&E and the immunohistochemical biomarker TFF3. This resource-intensive clinical workflow limits large-scale screening in the at-risk population. To improve screening capacity, we propose a deep learning approach for detecting Barrett's from routinely stained H&E slides. The approach solely relies on diagnostic labels, eliminating the need for expensive localized expert annotations. We train and independently validate our approach on two clinical trial datasets, totaling 1866 patients. We achieve 91.4% and 87.3% AUROCs on discovery and external test datasets for the H&E model, comparable to the TFF3 model. Our proposed semi-automated clinical workflow can reduce pathologists' workload to 48% without sacrificing diagnostic performance, enabling pathologists to prioritize high risk cases.

Description

Acknowledgements: The BEST2 study was funded by program grants from Cancer Research UK (BEST2 grant number C14478/A12088) and was supported by NIHR infrastructure for the Biomedical Research Center in Cambridge. The DELTA study was funded by an Innovate UK grant (grant number 41162). Innovate UK, University of Cambridge and Cambridge University Hospitals NHS Trust had no role in the design and conduct of the study; in the collection, analysis and interpretation of the data; or in the preparation, review or approval of the manuscript. We would like to thank Rebecca C Fitzgerald for making this data available, the pathologists from Addenbrookes Hospital, Cambridge, UK for their work scoring the BEST2 slides, and the pathologists at Cyted Ltd for scoring the DELTA slides. We would like to extend our thanks to Sophie Ghazal for support that was instrumental in laying the foundation for the study. We also thank Hannah Richardson for guidance offered as part the compliance review of the datasets used in this study. We thank Melissa Bristow for helping in maintaining the open-source repository, and Fernando Pérez García for helping with slide visualization tools. This work was funded by Microsoft Research Ltd (Cambridge, UK).

Keywords

Humans, Barrett Esophagus, Deep Learning, Esophageal Neoplasms, Adenocarcinoma, Metaplasia

Journal Title

Nat Commun

Conference Name

Journal ISSN

2041-1723
2041-1723

Volume Title

15

Publisher

Springer Science and Business Media LLC