Accurate prediction of histological grading of intraductal papillary mucinous neoplasia using deep learning
Endoscopy · 01.05.2023 · 55 (05) · 415-422
DOI: 10.1055/a-1971-1274 ↗Kurz gefasst
Fachartikel in der Zeitschrift Endoscopy, erschienen am 01.05.2023. Verfasst von Dominik Schulz und 11 weiteren Autor:innen. Beteiligte Einrichtung: Klinik und Poliklinik für Innere Medizin II, Klinikum rechts der Isar, Technical University of Munich, Munich, Germany. Hauptthemen laut MEDLINE: Carcinoma, Pancreatic Ductal, Deep Learning, Pancreatic Intraductal Neoplasms, Adenocarcinoma, Mucinous. Die Arbeit wurde bisher 23-mal zitiert (Crossref) und nennt 34 Literaturangaben. Gefördert von: Deutsche Krebshilfe, Deutsche Forschungsgemeinschaft.
Abstract (Originalsprache)
Background
Risk stratification and recommendation for surgery for intraductal papillary mucinous neoplasm (IPMN) are currently based on consensus guidelines. Risk stratification from presurgery histology is only potentially decisive owing to the low sensitivity of fine-needle aspiration. In this study, we developed and validated a deep learning-based method to distinguish between IPMN with low grade dysplasia and IPMN with high grade dysplasia/invasive carcinoma using endoscopic ultrasound (EUS) images.
Methods
For model training, we acquired a total of 3355 EUS images from 43 patients who underwent pancreatectomy from March 2015 to August 2021. All patients had histologically proven IPMN. We used transfer learning to fine-tune a convolutional neural network and to classify “low grade IPMN” from “high grade IPMN/invasive carcinoma.” Our test set consisted of 1823 images from 27 patients, recruiting 11 patients retrospectively, 7 patients prospectively, and 9 patients externally. We compared our results with the prediction based on international consensus guidelines.
Results
Our approach could classify low grade from high grade/invasive carcinoma in the test set with an accuracy of 99.6 % (95 %CI 99.5 %–99.9 %). Our deep learning model achieved superior accuracy in prediction of the histological outcome compared with any individual guideline, which have accuracies between 51.8 % (95 %CI 31.9 %–71.3 %) and 70.4 % (95 %CI 49.8–86.2).
Conclusion
This pilot study demonstrated that deep learning in IPMN-EUS images can predict the histological outcome with high accuracy.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Klinik und Poliklinik für Innere Medizin II, Klinikum rechts der Isar, Technical University of Munich, Munich, Germany.
- Klinik für Innere Medizin II, Universitätsklinikum Freiburg, Freiburg, Germany.
- Klinik und Poliklinik für Chirurgie, Klinikum rechts der Isar, Technical University of Munich, Munich, Germany.
- German Cancer Consortium (DKTK), Partner Site Munich, Munich, Germany.
ORCID-Profile: Dominik Schulz, Julius Mueller
Förderung
- Deutsche Krebshilfe
- Deutsche Forschungsgemeinschaft
Zitieren
APA
Schulz, D., Heilmaier, M., Phillip, V., Treiber, M., Mayr, U., Lahmer, T., Mueller, J., Demir, I. E., Friess, H., Reichert, M., Schmid, R. M., & Abdelhafez, M. (2023). Accurate prediction of histological grading of intraductal papillary mucinous neoplasia using deep learning. Endoscopy, 55(05), 415-422. https://doi.org/10.1055/a-1971-1274
Vancouver
Schulz D, Heilmaier M, Phillip V, Treiber M, Mayr U, Lahmer T, et al. Accurate prediction of histological grading of intraductal papillary mucinous neoplasia using deep learning. Endoscopy. 2023;55(05):415-422. doi:10.1055/a-1971-1274
BibTeX
@article{Schulz2023,
author = {Schulz, Dominik and Heilmaier, Markus and Phillip, Veit and Treiber, Matthias and Mayr, Ulrich and Lahmer, Tobias and Mueller, Julius and Demir, Ihsan Ekin and Friess, Helmut and Reichert, Maximilian and Schmid, Roland M and Abdelhafez, Mohamed},
title = {Accurate prediction of histological grading of intraductal papillary mucinous neoplasia using deep learning},
journal = {Endoscopy},
year = {2023},
volume = {55},
number = {05},
pages = {415--422},
doi = {10.1055/a-1971-1274}
}
EUVIMED