Endoscopy-Driven Pretraining for Classification of Dysplasia in Barrett’s Esophagus with Endoscopic Narrow-Band Imaging Zoom Videos
Source abstract
Abstract
Endoscopic diagnosis of early neoplasia in Barrett’s Esophagus is generally a two-step process of primary detection in overview, followed by detailed inspection of any visible abnormalities using Narrow Band Imaging (NBI). However, endoscopists struggle with evaluating NBI-zoom imagery of subtle abnormalities. In this work, we propose the first results of a deep learning system for the characterization of NBI-zoom imagery of Barrett’s Esophagus with an accuracy, sensitivity, and specificity of 83.6%, 83.1%, and 84.0%, respectively. We also show that endoscopy-driven pretraining outperforms two models, one without pretraining as well as a model with ImageNet initialization. The final model outperforms absence of pretraining by approximately 10% and the performance is 2% higher in terms of accuracy compared to ImageNet pretraining. Furthermore, the practical deployment of our model is not hampered by ImageNet licensing, thereby paving the way for clinical application.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Esophageal Cancer Research and Treatment · Esophageal and GI Pathology · Gastric Cancer Management and Outcomes
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Electrical Engineering, Video Coding and Architectures, Eindhoven University of Technology, 5612 AZ Eindhoven, Noord-Brabant, The Netherlands
Department of Gastroenterology and Hepatology, Amsterdam UMC, University of Amsterdam, 1105 AZ Amsterdam, Noord-Holland, The Netherlands
Department of Gastroenterology and Hepatology, Catharina Hospital, 5623 EJ Eindhoven, Noord-Brabant, The Netherlands
Department of Gastroenterology and Hepatology, Karolinksa University Hospital, SE-171 76 Solna, Stockholm, Sweden
Funding
Nederlandse Organisatie voor Wetenschappelijk Onderzoek
References (24)
Show source-provided references
1. Endoscopic management of Barrett’s esophagus: European Society of Gastrointestinal Endoscopy (ESGE) position statement10.1055/s-0042-122140 ↗
2. Long-term efficacy and safety of endoscopic resection for patients with mucosal adenocarcinoma of the esophagus10.1053/j.gastro.2013.11.006 ↗
3. Cost-effectiveness of endoscopic surveillance of non-dysplastic Barrett’s esophagus10.1016/j.gie.2013.07.046 ↗
4. Detection and classification of the mucosal and vascular patterns (mucosal morphology) in Barrett’s esophagus by using narrow band imaging10.1016/j.gie.2005.11.049 ↗
5. Observer agreement in the assessment of narrowband imaging system surface patterns in Barrett’s esophagus: A multicenter study10.1055/s-0030-1256631 ↗
6. Development and validation of a classification system to identify high-grade dysplasia and esophageal adenocarcinoma in Barrett’s esophagus using narrow-band imaging10.1053/j.gastro.2015.11.037 ↗
7. Real-time differentiation of adenomatous and hyperplastic diminutive colorectal polyps during analysis of unaltered videos of standard colonoscopy using a deep learning model10.1136/gutjnl-2017-314547 ↗
8. Accurate classification of diminutive colorectal polyps using computer-aided analysis10.1053/j.gastro.2017.10.010 ↗
9. The Argos project: The development of a computer-aided detection system to improve detection of Barrett’s neoplasia on white light endoscopy10.1177/2050640619837443 ↗
10. Deep-Learning System Detects Neoplasia in Patients With Barrett’s Esophagus With Higher Accuracy Than Endoscopists in a Multi-Step Training and Validation Study with Benchmarking10.1053/j.gastro.2019.11.030 ↗
11. Computer-aided colorectal tumor classification in NBI endoscopy using local features10.1016/j.media.2012.08.003 ↗
12. Automatic detection and classification of colorectal polyps by transferring low-level CNN features from nonmedical domain10.1109/JBHI.2016.2635662 ↗
13. van der Putten, J., Wildeboer, R., de Groof, J., van Sloun, R., Struyvenberg, M., van der Sommen, F., Zinger, S., Curvers, W., Schoon, E., and Bergman, J. (2019, January 8–11). Deep Learning Biopsy Marking of Early Neoplasia in Barrett’s Esophagus by Combining WLE and BLI Modalities. Proceedings of the 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019), Venice, Italy.10.1109/ISBI.2019.8759431 ↗
14. Computer-aided diagnosis using deep learning in the evaluation of early oesophageal adenocarcinoma10.1136/gutjnl-2018-317573 ↗
15. Van der Putten, J., de Groof, J., van der Sommen, F., Struyvenberg, M., Zinger, S., Curvers, W., Schoon, E., Bergman, J., and de With, P.H.N. (2019). Pseudo-labeled Bootstrapping and Multi-stage Transfer Learning for the Classification and Localization of Dysplasia in Barrett’s Esophagus. International Workshop on Machine Learning in Medical Imaging, Springer.10.1007/978-3-030-32692-0_20 ↗
16. The ARGOS project: First deep learning algorithm for detection of Barrett’s neoplasia outperforms conventional computer aided detection systems in a multi-step training and external validation study10.1016/j.gie.2019.04.094 ↗
17. He, K., Zhang, X., Ren, S., and Sun, J. (July, January 26). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.
18. Ioffe, S., and Szegedy, C. (2015). Batch normalization: Accelerating deep network training by reducing internal covariate shift. arXiv.
19. Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., and Fei-Fei, L. (2009, January 20–25). Imagenet: A large-scale hierarchical image database. Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, USA.10.1109/CVPR.2009.5206848 ↗
20. Kingma, D.P., and Ba, J. (2014). Adam: A method for stochastic optimization. arXiv.
21. Reddi, S.J., Kale, S., and Kumar, S. (2019). On the convergence of adam and beyond. arXiv.
22. Smith, L.N. (2017, January 24–31). Cyclical learning rates for training neural networks. Proceedings of the 2017 IEEE Winter Conference on Applications of Computer Vision (WACV), Santa Rosa, CA, USA.10.1109/WACV.2017.58 ↗
23. Critical values and probability levels for the Wilcoxon rank sum test and the Wilcoxon signed rank test
24. Van der Putten, J., de Groof, J., van der Sommen, F., Struyvenberg, M., Zinger, S., Curvers, W., Schoon, E., Bergman, J., and de With, P.H. (2019, January 22–25). Informative frame classification of endoscopic videos using convolutional neural networks and hidden Markov models. Proceedings of the 2019 IEEE International Conference on Image Processing (ICIP), Taipei, Taiwan.10.1109/ICIP.2019.8802947 ↗