EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-SA 4.0

A deep learning–based system to identify originating mural layer of upper gastrointestinal submucosal tumors under EUS

Xun Li, Chenxia Zhang, Liwen Yao, Jun Zhang, Kun Zhang, Hui Feng, Honggang Yu

Endoscopic Ultrasound · 2023-11 · 12 (6) · 465-471

DOI: 10.1097/eus.0000000000000029 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Endoscopic Ultrasound, online erschienen am 22.12.2023. Verfasst von Xun Li und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Gastroenterology, Renmin Hospital of Wuhan University, Wuhan, Hubei Province, China. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 61 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-SA 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background and Objective EUS is the most accurate procedure to determine the originating mural layer and subsequently select the treatment of submucosal tumors (SMTs). However, it requires superb technical and cognitive skills. In this study, we propose a system named SMT Master to determine the originating mural layer of SMTs under EUS. Materials and Methods We developed 3 models: deep convolutional neural network (DCNN) 1 for lesion segmentation, DCNN2 for mural layer segmentation, and DCNN3 for the originating mural layer classification. A total of 2721 EUS images from 201 patients were used to train the 3 models. We validated our model internally and externally using 283 images from 26 patients and 172 images from 26 patients, respectively. We applied 368 images from 30 patients for the man-machine contest and used 30 video clips to test the originating mural layer classification.

Results

In the originating mural layer classification task, DCNN3 achieved a classification accuracy of 84.43% and 80.68% at internal and external validations, respectively. In the video test, the accuracy was 80.00%. DCNN1 achieved Dice coefficients of 0.956 and 0.776 for lesion segmentation at internal and external validations, respectively, whereas DCNN2 achieved Dice coefficients of 0.820 and 0.740 at internal and external validations, respectively. The system achieved 90.00% accuracy in classification, which is comparable with that of EUS experts.

Conclusions

Our proposed system has the potential to solve difficulties in determining the originating mural layer of SMTs in EUS procedures, which relieves the EUS learning pressure of physicians.

Schlagwörter

Stichwörter der Autor:innen

Deep learningEUSSubmucosal tumors

Autor:innen und Institutionen

  • Department of Gastroenterology, Renmin Hospital of Wuhan University, Wuhan, Hubei Province, China.
  • Key Laboratory of Hubei Province for Digestive System Disease, Renmin Hospital of Wuhan University, Wuhan, Hubei Province, China.
  • Hubei Provincial Clinical Research Center for Digestive Disease Minimally Invasive Incision, Renmin Hospital of Wuhan University, Wuhan, Hubei Province, China.
  • Wuhan Union Hospital, Huazhong University of Science and Technology, Wuhan, Hubei Province, China.
  • Information center, Renmin Hospital of Wuhan University, Wuhan, Hubei Province, China.

Zitieren

APA

Li, X., Zhang, C., Yao, L., Zhang, J., Zhang, K., Feng, H., & Yu, H. (2023). A deep learning–based system to identify originating mural layer of upper gastrointestinal submucosal tumors under EUS. Endoscopic Ultrasound, 12(6), 465-471. https://doi.org/10.1097/eus.0000000000000029

Vancouver

Li X, Zhang C, Yao L, Zhang J, Zhang K, Feng H, et al. A deep learning–based system to identify originating mural layer of upper gastrointestinal submucosal tumors under EUS. Endoscopic Ultrasound. 2023;12(6):465-471. doi:10.1097/eus.0000000000000029

BibTeX
@article{Li2023,
  author = {Li, Xun and Zhang, Chenxia and Yao, Liwen and Zhang, Jun and Zhang, Kun and Feng, Hui and Yu, Honggang},
  title = {A deep learning–based system to identify originating mural layer of upper gastrointestinal submucosal tumors under EUS},
  journal = {Endoscopic Ultrasound},
  year = {2023},
  volume = {12},
  number = {6},
  pages = {465--471},
  doi = {10.1097/eus.0000000000000029}
}
Kostenlos starten →