An artificial intelligence algorithm is highly accurate for detecting endoscopic features of eosinophilic esophagitis - PubMed (original) (raw)

. 2022 Jul 1;12(1):11115.

doi: 10.1038/s41598-022-14605-z.

Robert Mendel # 2 3, Caroline Barrett 4, Hans Kiesl 5, David Rauber 2, Tobias Rückert 2, Lisa Kraus 1, Jakob Heinkele 1, Christine Dhillon 6, Bianca Grosser 6, Friederike Prinz 1, Julia Wanzl 1, Carola Fleischmann 1, Sandra Nagl 1, Elisabeth Schnoy 1, Jakob Schlottmann 1, Evan S Dellon 4, Helmut Messmann 1, Christoph Palm 2 3, Alanna Ebigbo 7

Affiliations

An artificial intelligence algorithm is highly accurate for detecting endoscopic features of eosinophilic esophagitis

Christoph Römmele et al. Sci Rep. 2022.

Abstract

The endoscopic features associated with eosinophilic esophagitis (EoE) may be missed during routine endoscopy. We aimed to develop and evaluate an Artificial Intelligence (AI) algorithm for detecting and quantifying the endoscopic features of EoE in white light images, supplemented by the EoE Endoscopic Reference Score (EREFS). An AI algorithm (AI-EoE) was constructed and trained to differentiate between EoE and normal esophagus using endoscopic white light images extracted from the database of the University Hospital Augsburg. In addition to binary classification, a second algorithm was trained with specific auxiliary branches for each EREFS feature (AI-EoE-EREFS). The AI algorithms were evaluated on an external data set from the University of North Carolina, Chapel Hill (UNC), and compared with the performance of human endoscopists with varying levels of experience. The overall sensitivity, specificity, and accuracy of AI-EoE were 0.93 for all measures, while the AUC was 0.986. With additional auxiliary branches for the EREFS categories, the AI algorithm (AI-EoE-EREFS) performance improved to 0.96, 0.94, 0.95, and 0.992 for sensitivity, specificity, accuracy, and AUC, respectively. AI-EoE and AI-EoE-EREFS performed significantly better than endoscopy beginners and senior fellows on the same set of images. An AI algorithm can be trained to detect and quantify endoscopic features of EoE with excellent performance scores. The addition of the EREFS criteria improved the performance of the AI algorithm, which performed significantly better than endoscopists with a lower or medium experience level.

© 2022. The Author(s).

PubMed Disclaimer

Conflict of interest statement

HM has received a research grant or has served as a consultant for: Apollo Endosurgery, Biogen, Boston Scientific, CDx Diagnostic, Cook Medical, CSL Behring, Dr. Falk Pharma, Endo Tools Therapeutics, Erbe, Fujifilm, Hitachi, Janssen-Cilag, Medwork, Norgine, Nutricia, Olympus, Ovesco Endoscopy, Servier Deutschland, US Endoscopy(Endoscopic companies); Amgen, Bayer, Dr. Falk Pharma, MSD, Novartis Olympus, Roche (Grants); Covidien, Dr. Falk Pharma, Olympus (Honorarium); Boston Scientific, CDx Diagnostics, Covidien, Erbe, Lumendi, Norgine, Olympus (Consultation fees); Stock shareholder: the other authors declare no competing interests.

Figures

Figure 1

Figure 1

Endoscopic white light images of eosinophilic esophagitis showing furrows, exudates, edema, and rings.

Figure 2

Figure 2

Endoscopic white light images of a normal esophagus.

Figure 3

Figure 3

ROC curves and AUC values of AI-EoE and AI-EREFS on the internal data set (InD).

Figure 4

Figure 4

ROC curves and AUC values of AI-EoE and AI-EoE-EREFS on the external data set (ExD) compared with human endoscopists who had varying experience levels.

Figure 5

Figure 5

Features detected on input images by AI-EoE-EREFS are highlighted using Gradient-based visualization (Grad-CAM): the top left image shows the original endoscopic image with furrows, exudates, and rings; in the top right image, furrows are highlighted, while in the bottom left and bottom right images, exudates, and rings are highlighted, respectively.

Similar articles

Cited by

References

    1. Kovačić M, Unić J, Mišak Z, et al. One-year outcomes in children with eosinophilic esophagitis. Esophagus. 2019;16:162–167. doi: 10.1007/s10388-018-0647-3. - DOI - PubMed
    1. Lucendo AJ, Molina-Infante J, Arias Á, et al. Guidelines on eosinophilic esophagitis: Evidence-based statements and recommendations for diagnosis and management in children and adults. United Eur. Gastroenterol. J. 2017;5:335–358. doi: 10.1177/2050640616689525. - DOI - PMC - PubMed
    1. Dellon ES, Hirano I. Epidemiology and natural history of eosinophilic esophagitis. Gastroenterology. 2018;154:319–32.e3. doi: 10.1053/j.gastro.2017.06.067. - DOI - PMC - PubMed
    1. Nielsen JA, Lager DJ, Lewin M, et al. The optimal number of biopsy fragments to establish a morphologic diagnosis of eosinophilic esophagitis. Am. J. Gastroenterol. 2014;109:515–520. doi: 10.1038/ajg.2013.463. - DOI - PubMed
    1. Dellon ES, Liacouras CA, Molina-Infante J, et al. Updated international consensus diagnostic criteria for eosinophilic esophagitis: Proceedings of the AGREE conference. Gastroenterology. 2018;155:1022–1033. doi: 10.1053/j.gastro.2018.07.009. - DOI - PMC - PubMed

Publication types

MeSH terms

LinkOut - more resources