Improving Automatic Endoscopic Stone Recognition Using a Multi-View Fusion Approach Enhanced with Two-Step Transfer Learning

Francisco Javier López-Tiro, Elias Villalvazo-Avila, Juan Pablo Betancur-Rengifo, Iván Reyes-Amezcua, Jacques Hubert, Gilberto Ochoa-Ruiz, Christian Daul

ICCVW 2023 pp. 4167-4174

doi:10.1109/ICCVW60793.2023.00450 /iccvw/2023/lopeztiro2023iccvw-improving/

Abstract

This contribution presents a deep-learning method for extracting and fusing image information acquired from different viewpoints, with the aim to produce more discriminant object features for the identification of the type of kidney stones seen in endoscopic images. The approach was specifically designed to mimic the morpho-constitutional analysis to visually classify kidney stones by jointly using surface and section images of kidney stone fragments. The model was further improved with a two-step transfer learning approach and by attention blocks to refine the learned feature maps. Deep feature fusion strategies improved the results of single view extraction backbone models by more than 6% in terms of accuracy of the kidney stones classification.

PDF ICCVW Semantic Scholar

Cite

Text

López-Tiro et al. "Improving Automatic Endoscopic Stone Recognition Using a Multi-View Fusion Approach Enhanced with Two-Step Transfer Learning." IEEE/CVF International Conference on Computer Vision Workshops, 2023. doi:10.1109/ICCVW60793.2023.00450

Markdown

[López-Tiro et al. "Improving Automatic Endoscopic Stone Recognition Using a Multi-View Fusion Approach Enhanced with Two-Step Transfer Learning." IEEE/CVF International Conference on Computer Vision Workshops, 2023.](https://mlanthology.org/iccvw/2023/lopeztiro2023iccvw-improving/) doi:10.1109/ICCVW60793.2023.00450

BibTeX

@inproceedings{lopeztiro2023iccvw-improving,
  title     = {{Improving Automatic Endoscopic Stone Recognition Using a Multi-View Fusion Approach Enhanced with Two-Step Transfer Learning}},
  author    = {López-Tiro, Francisco Javier and Villalvazo-Avila, Elias and Betancur-Rengifo, Juan Pablo and Reyes-Amezcua, Iván and Hubert, Jacques and Ochoa-Ruiz, Gilberto and Daul, Christian},
  booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
  year      = {2023},
  pages     = {4167-4174},
  doi       = {10.1109/ICCVW60793.2023.00450},
  url       = {https://mlanthology.org/iccvw/2023/lopeztiro2023iccvw-improving/}
}