Semantic Parsing of Colonoscopy Videos with Multi-Label Temporal Networks

Ori Kelner, Or Weinstein, Ehud Rivlin, Roman Goldenberg

ICCVW 2023 pp. 2591-2598

doi:10.1109/ICCVW60793.2023.00274 /iccvw/2023/kelner2023iccvw-semantic/

Abstract

Following the successful debut of polyp detection and characterization, more advanced automation tools are being developed for colonoscopy. The new automation tasks, such as quality metrics or report generation, require understanding of the procedure flow that includes activities, events, anatomical landmarks, etc. In this work we present a method for automatic semantic parsing of colonoscopy videos. The method uses a novel DL multi-label temporal segmentation model trained in supervised and unsupervised regimes. We evaluate the accuracy of the method on a test set of over 300 annotated colonoscopy videos, and use ablation to explore the relative importance of various method’s components.

PDF ICCVW Semantic Scholar

Cite

Text

Kelner et al. "Semantic Parsing of Colonoscopy Videos with Multi-Label Temporal Networks." IEEE/CVF International Conference on Computer Vision Workshops, 2023. doi:10.1109/ICCVW60793.2023.00274

Markdown

[Kelner et al. "Semantic Parsing of Colonoscopy Videos with Multi-Label Temporal Networks." IEEE/CVF International Conference on Computer Vision Workshops, 2023.](https://mlanthology.org/iccvw/2023/kelner2023iccvw-semantic/) doi:10.1109/ICCVW60793.2023.00274

BibTeX

@inproceedings{kelner2023iccvw-semantic,
  title     = {{Semantic Parsing of Colonoscopy Videos with Multi-Label Temporal Networks}},
  author    = {Kelner, Ori and Weinstein, Or and Rivlin, Ehud and Goldenberg, Roman},
  booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
  year      = {2023},
  pages     = {2591-2598},
  doi       = {10.1109/ICCVW60793.2023.00274},
  url       = {https://mlanthology.org/iccvw/2023/kelner2023iccvw-semantic/}
}