Improving Colonoscopy Lesion Classification Using Semi-Supervised Deep Learning
| Autor: | Simran Jit, Saowanee Ngamruengphong, Nicholas J. Durr, Mayank Golhar, MirMilad Pourmousavi Khoshknab, Taylor L. Bobrow |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
FOS: Computer and information sciences
General Computer Science Computer science domain adaptation Computer Vision and Pattern Recognition (cs.CV) Computer Science - Computer Vision and Pattern Recognition Colonoscopy Economic shortage semi-supervised Machine learning computer.software_genre Article 030218 nuclear medicine & medical imaging 03 medical and health sciences 0302 clinical medicine Medical imaging medicine unsupervised General Materials Science endoscopy jigsaw lesion classification medicine.diagnostic_test business.industry Deep learning Supervised learning General Engineering deep learning ComputingMethodologies_PATTERNRECOGNITION Task analysis 030211 gastroenterology & hepatology lcsh:Electrical engineering. Electronics. Nuclear engineering Artificial intelligence business lcsh:TK1-9971 computer out-of-distribution detection |
| Zdroj: | IEEE access : practical innovations, open solutions IEEE Access, Vol 9, Pp 631-640 (2021) |
| ISSN: | 2169-3536 |
| Popis: | While data-driven approaches excel at many image analysis tasks, the performance of these approaches is often limited by a shortage of annotated data available for training. Recent work in semi-supervised learning has shown that meaningful representations of images can be obtained from training with large quantities of unlabeled data, and that these representations can improve the performance of supervised tasks. Here, we demonstrate that an unsupervised jigsaw learning task, in combination with supervised training, results in up to a 9.8% improvement in correctly classifying lesions in colonoscopy images when compared to a fully-supervised baseline. We additionally benchmark improvements in domain adaptation and out-of-distribution detection, and demonstrate that semi-supervised learning outperforms supervised learning in both cases. In colonoscopy applications, these metrics are important given the skill required for endoscopic assessment of lesions, the wide variety of endoscopy systems in use, and the homogeneity that is typical of labeled datasets. 10 pages, 5 figures |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|