Endoscopy 2021; 53(09): 878-883
DOI: 10.1055/a-1311-8570
Original article

Endoscopic prediction of submucosal invasion in Barrett’s cancer with the use of artificial intelligence: a pilot study

Alanna Ebigbo*
 1  III Medizinische Klinik, Universitätsklinikum Augsburg, Augsburg Germany
,
Robert Mendel*
 2  Regensburg Medical Image Computing (ReMIC), Ostbayerische Technische Hochschule Regensburg (OTH Regensburg), Regensburg Germany
 3  Regensburg Center of Health Sciences and Technology (RCHST), OTH Regensburg, Regensburg, Germany
,
Tobias Rückert
 2  Regensburg Medical Image Computing (ReMIC), Ostbayerische Technische Hochschule Regensburg (OTH Regensburg), Regensburg Germany
,
Laurin Schuster
 2  Regensburg Medical Image Computing (ReMIC), Ostbayerische Technische Hochschule Regensburg (OTH Regensburg), Regensburg Germany
,
Andreas Probst
 1  III Medizinische Klinik, Universitätsklinikum Augsburg, Augsburg Germany
,
Johannes Manzeneder
 1  III Medizinische Klinik, Universitätsklinikum Augsburg, Augsburg Germany
,
Friederike Prinz
 1  III Medizinische Klinik, Universitätsklinikum Augsburg, Augsburg Germany
,
Matthias Mende
 4  Gastroenterology, Sana Klinikum Lichtenberg, Berlin, Germany
,
Ingo Steinbrück
 5  Department of Gastroenterology, Hepatology and Interventional Endoscopy, Asklepios Klinik Barmbek, Hamburg, Germany
,
Siegbert Faiss
 4  Gastroenterology, Sana Klinikum Lichtenberg, Berlin, Germany
,
David Rauber
 2  Regensburg Medical Image Computing (ReMIC), Ostbayerische Technische Hochschule Regensburg (OTH Regensburg), Regensburg Germany
 6  Regensburg Center of Biomedical Engineering (RCBE), OTH Regensburg and Regensburg University, Regensburg, Germany
,
Luis A. de Souza Jr.
 2  Regensburg Medical Image Computing (ReMIC), Ostbayerische Technische Hochschule Regensburg (OTH Regensburg), Regensburg Germany
 7  Department of Computing, São Paulo State University, São Paulo, Brazil
,
João P. Papa
 7  Department of Computing, São Paulo State University, São Paulo, Brazil
,
 8  Cliniques Universitaires St-Luc, Université Catholique de Louvain, Brussels, Belgium
,
Tsuneo Oyama
 9  Saku Central Hospital Advanced Care Center, Nagano, Japan
,
Akiko Takahashi
 9  Saku Central Hospital Advanced Care Center, Nagano, Japan
,
Stefan Seewald
10  GastroZentrum, Klinik Hirslanden, Zurich, Switzerland
,
Prateek Sharma
11  Department of Gastroenterology and Hepatology, Veterans Affairs Medical Center and University of Kansas School of Medicine, Kansas City, Missouri, United States
,
Michael F. Byrne
12  Division of Gastroenterology, Vancouver General Hospital, University of British Columbia, Vancouver, British Columbia, Canada
,
Christoph Palm
 2  Regensburg Medical Image Computing (ReMIC), Ostbayerische Technische Hochschule Regensburg (OTH Regensburg), Regensburg Germany
 3  Regensburg Center of Health Sciences and Technology (RCHST), OTH Regensburg, Regensburg, Germany
 6  Regensburg Center of Biomedical Engineering (RCBE), OTH Regensburg and Regensburg University, Regensburg, Germany
,
Helmut Messmann
 1  III Medizinische Klinik, Universitätsklinikum Augsburg, Augsburg Germany
› Author Affiliations

Abstract

Background The accurate differentiation between T1a and T1b Barrett’s-related cancer has both therapeutic and prognostic implications but is challenging even for experienced physicians. We trained an artificial intelligence (AI) system on the basis of deep artificial neural networks (deep learning) to differentiate between T1a and T1b Barrett’s cancer on white-light images.

Methods Endoscopic images from three tertiary care centers in Germany were collected retrospectively. A deep learning system was trained and tested using the principles of cross validation. A total of 230 white-light endoscopic images (108 T1a and 122 T1b) were evaluated using the AI system. For comparison, the images were also classified by experts specialized in endoscopic diagnosis and treatment of Barrett’s cancer.

Results The sensitivity, specificity, F1 score, and accuracy of the AI system in the differentiation between T1a and T1b cancer lesions was 0.77, 0.64, 0.74, and 0.71, respectively. There was no statistically significant difference between the performance of the AI system and that of experts, who showed sensitivity, specificity, F1, and accuracy of 0.63, 0.78, 0.67, and 0.70, respectively.

Conclusion This pilot study demonstrates the first multicenter application of an AI-based system in the prediction of submucosal invasion in endoscopic images of Barrett’s cancer. AI scored equally to international experts in the field, but more work is necessary to improve the system and apply it to video sequences and real-life settings. Nevertheless, the correct prediction of submucosal invasion in Barrett’s cancer remains challenging for both experts and AI.

* These authors contributed equally to this work.




Publication History

Received: 01 June 2020

Accepted: 16 November 2020

Publication Date:
16 November 2020 (online)

© 2020. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany