Vision transformer distillation for enhanced gastrointestinal abnormality recognition in wireless capsule endoscopy images
We proposed a deep learning approach utilizing CNNs and a ViT with KD to effectively classify gastrointestinal diseases in WCE images. It demonstrates promising performance on public datasets, distinguishing normal from abnormal regions and bleeding from non-bleeding cases while offering optimal computational efficiency compared with existing methods, making it suitable for GI disease applications.
Authors
- Zakaria Kerkaou
- Cosimo Distante (ORCID: https://orcid.org/0000-0002-1073-2390)
- Nikolaos Papachrysos (ORCID: https://orcid.org/0000-0002-2372-2554)
- Lahcen Koutti (ORCID: https://orcid.org/0000-0002-4274-3414)
- Anass Garbaz (ORCID: https://orcid.org/0000-0002-8526-3955)
- Yassine Oukdach (ORCID: https://orcid.org/0009-0000-9748-2875)
- Thomas de Lange (ORCID: https://orcid.org/0000-0003-3989-7487)
- Mohamed El Ansari (ORCID: https://orcid.org/0000-0001-5394-9066)
- Ahmed Fouad El Ouafdi
Institutions
- Université Ibn Zohr (MA)
- Institute of Applied Science and Intelligent Systems (IT)
- Université Moulay Ismail de Meknes (MA)
- University of Gothenburg (SE)
Publication Details
- Journal
- Journal of Medical Imaging
- Published
- 2025-02-05
- DOI
- https://doi.org/10.1117/1.jmi.12.1.014505
- Citations
- 3
- Primary Topic
- Gastrointestinal Bleeding Diagnosis and Treatment
- Type
- article
- Field-Weighted Citation Impact
- 5.43