Assisted documentation as a new focus for artificial intelligence in endoscopy: the precedent of reliable withdrawal time and image reporting
Endoscopy · 01.12.2023 · 55 (12) · 1118-1123
DOI: 10.1055/a-2122-1671 ↗Kurz gefasst
Fachartikel in der Zeitschrift Endoscopy, erschienen am 01.12.2023. Verfasst von Thomas J Lux und 9 weiteren Autor:innen. Beteiligte Einrichtung: Interventional and Experimental Endoscopy (InExEn), Internal Medicine II, University Hospital Würzburg, Würzburg, Germany. Hauptthemen laut MEDLINE: Artificial Intelligence, Endoscopy, Gastrointestinal. Die Arbeit wurde bisher 33-mal zitiert (Crossref) und nennt 17 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Baden-Württemberg Stiftung, Eva Mayr-Stihl Stiftung, Dieter von Holtzbrinck Stiftung GmbH, Stuttgart, Germany.
Abstract (Originalsprache)
Background
Reliable documentation is essential for maintaining quality standards in endoscopy; however, in clinical practice, report quality varies. We developed an artificial intelligence (AI)-based prototype for the measurement of withdrawal and intervention times, and automatic photodocumentation.
Method
A multiclass deep learning algorithm distinguishing different endoscopic image content was trained with 10 557 images (1300 examinations, nine centers, four processors). Consecutively, the algorithm was used to calculate withdrawal time (AI prediction) and extract relevant images. Validation was performed on 100 colonoscopy videos (five centers). The reported and AI-predicted withdrawal times were compared with video-based measurement; photodocumentation was compared for documented polypectomies.
Results
Video-based measurement in 100 colonoscopies revealed a median absolute difference of 2.0 minutes between the measured and reported withdrawal times, compared with 0.4 minutes for AI predictions. The original photodocumentation represented the cecum in 88 examinations compared with 98/100 examinations for the AI-generated documentation. For 39/104 polypectomies, the examiners’ photographs included the instrument, compared with 68 for the AI images. Lastly, we demonstrated real-time capability (10 colonoscopies).
Conclusion
Our AI system calculates withdrawal time, provides an image report, and is real-time ready. After further validation, the system may improve standardized reporting, while decreasing the workload created by routine documentation.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Interventional and Experimental Endoscopy (InExEn), Internal Medicine II, University Hospital Würzburg, Würzburg, Germany.
- Department of Internal Medicine and Gastroenterology, Katharinenhospital, Stuttgart, Germany.
- Department for Interdisciplinary Endoscopy; Department of Internal Medicine I; and Department of Computational Neuroscience, University Hospital Hamburg - Eppendorf, Hamburg, Germany.
ORCID-Profile: Thomas J Lux, Alexander Hann
Förderung
- Baden-Württemberg Stiftung
- Eva Mayr-Stihl Stiftung
- Dieter von Holtzbrinck Stiftung GmbH, Stuttgart, Germany
- Fischerwerke GmbH & Co. KG, Waldachtal, Germany
Zitieren
APA
Lux, T. J., Saßmannshausen, Z., Kafetzis, I., Sodmann, P., Herold, K., Sudarevic, B., Schmitz, R., Zoller, W. G., Meining, A., & Hann, A. (2023). Assisted documentation as a new focus for artificial intelligence in endoscopy: the precedent of reliable withdrawal time and image reporting. Endoscopy, 55(12), 1118-1123. https://doi.org/10.1055/a-2122-1671
Vancouver
Lux TJ, Saßmannshausen Z, Kafetzis I, Sodmann P, Herold K, Sudarevic B, et al. Assisted documentation as a new focus for artificial intelligence in endoscopy: the precedent of reliable withdrawal time and image reporting. Endoscopy. 2023;55(12):1118-1123. doi:10.1055/a-2122-1671
BibTeX
@article{Lux2023,
author = {Lux, Thomas J and Saßmannshausen, Zita and Kafetzis, Ioannis and Sodmann, Philipp and Herold, Katja and Sudarevic, Boban and Schmitz, Rüdiger and Zoller, Wolfram G and Meining, Alexander and Hann, Alexander},
title = {Assisted documentation as a new focus for artificial intelligence in endoscopy: the precedent of reliable withdrawal time and image reporting},
journal = {Endoscopy},
year = {2023},
volume = {55},
number = {12},
pages = {1118--1123},
doi = {10.1055/a-2122-1671}
}
EUVIMED