Proceeding of the Faculty of technical Sciences, Novi Sad

Electrotechnical and Computer Engineering

Vol. 41 No. 02 (2026): Proceedings of the Faculty of Technical Sciences

Automatic detection and classification of pathology in gastroscopic and colonoscopic images

  • Светлана Крунић
DOI:
https://doi.org/10.24867/34BE11Krunic
Submitted
February 12, 2026
Published
2026-03-09

Abstract

The aim of this work is to design and train an automatic system for the analysis of gastroscopic and colonoscopic images in the visible spectrum for the detection of the presence of pathomorphological changes and their categorization. The system is based on modern deep learning methods and is capable of autonomously assessing, based on images obtained during standard diagnostic endoscopic procedures, whether pathomorphological changes are present, to which category they belong, and presenting the results as probabilities of pathology presence (abnormal findings) and probabilities of different disease categories. The practical goal was to provide diagnostic support to physicians during the diagnostic process.

References

  1. [1] Borgli, H., Thambawita, V., Smedsrud, P.H. et al. HyperKvasir, a comprehensive multi-class image and video dataset for gastrointestinal endoscopy. Sci Data 7, 283 (2020). https://doi.org/10.1038/s41597-020-00622-y
  2. [2] Oquab, Maxime, et al. "Dinov2: Learning robust visual features without supervision." arXiv preprint arXiv:2304.07193 (2023).
  3. [3] Loshchilov, Ilya, and Frank Hutter. "Decoupled weight decay regularization." arXiv preprint arXiv:1711.05101 (2017).
  4. [4] Srivastava, Nitish, et al. "Dropout: A Simple Way to Prevent Neural Networks from Overfitting." Journal of Machine Learning Research 15.1 (2014): 1929–1958.
  5. [5] Siméoni, Oriane & Vo, Van Huy & Seitzer, Maximilian & Baldassarre, Federico & Oquab, Maxime & Jose, Cijo & Khalidov, Vasil & Szafraniec, Marc & Yi, Seungeun & Ramamonjisoa, Michaël & Massa, Francisco & Haziza, Daniel & Wehrstedt, Luca & Wang, Jianyuan & Darcet, Timothée & Moutakanni, Théo & Sentana, Leonel & Roberts, Claire & Vedaldi, Andrea & Bojanowski, Piotr. (2025). DINOv3. 10.48550/arXiv.2508.10104.
  6. [6] D. M. Hasan, A. H. S. A. Al-Khafaji, and R. A. Al-Rizzo, “EndoNet: A Multilevel Deep Feature Fusion Framework for Gastrointestinal Disease Classification,” Diagnostics, vol. 13, no. 4, pp. 671–686, 2023. DOI: 10.3390/diagnostics13040671.
  7. [7] A. K. Dutta, M. Biswas, and S. K. Das, “A triple-pronged approach for ulcerative colitis severity classification using multimodal, meta, and transformer-based learning,” Computers in Biology and Medicine, vol. 171, 107950, 2024. DOI: 10.1016/j.compbiomed.2024.107950.
  8. [8] N. Ramamurthy, S. Rajan, and P. K. Reddy, “A New Approach for Gastrointestinal Tract Findings Detection and Classification Deep Learning-Based Hybrid Stacking Ensemble Models,” Computers in Biology and Medicine, vol. 157, 106723, 2023. DOI: 10.1016/j.compbiomed.2023.106723