Enhancing Endoscopic Artefact Classification With GAN‐Based Synthetic Images Using Artefact Localisation Loss
Authors:
Open Access
- 1 January 2026
- journal article
- Published by Institution of Engineering and Technology (IET) in Electronics Letters
- Vol. 62 (1)
Abstract
Accurate artefact classification is essential for reliable diagnosis and computer‐aided analysis in endoscopy, as visual artefacts can severely degrade image quality. However, endoscopic artefact classification is often hindered by limited and imbalanced datasets. This letter introduces an artefact localisation loss (ALL) that guides StyleGAN2 with adaptive discriminator augmentation (StyleGAN2‐ADA) to focus on artefact regions, thereby improving the realism of synthetic images generated from small datasets. The synthetic images were then used for data augmentation, which significantly enhanced classification performance, particularly for underrepresented artefact classes. Experiments on in‐house and endoscopy artefact detection challenge (EAD2020) datasets achieved lower FID scores and showed consistent improvements in F1‐scores, demonstrating the effectiveness of the proposed ALL and synthetic augmentation for robust artefact classification.Keywords
This publication has 21 references indexed in Scilit:
- MobileNetV4: Universal Models for the Mobile Ecosystem2024
- Data-augmented landslide displacement prediction using generative adversarial networkJournal of Rock Mechanics and Geotechnical Engineering, 2024
- Medical image data augmentation: techniques, comparisons and interpretationsArtificial Intelligence Review, 2023
- Data augmentation for medical imaging: A systematic literature reviewComputers in Biology and Medicine, 2022
- Improving Artifact Detection in Endoscopic Video Frames Using Deep Learning Techniques2022
- Endoscopy Artefact Detection by Deep Transfer Learning of Baseline ModelsJournal of Digital Imaging, 2022
- Generative Adversarial Networks in Medical Image augmentation: A reviewComputers in Biology and Medicine, 2022
- Transfer learning for informative-frame selection in laryngoscopic videos through learned featuresMedical & Biological Engineering & Computing, 2020
- Learning-based classification of informative laryngoscopic framesComputer Methods and Programs in Biomedicine, 2018
- Automatic Assessment of Image Informativeness in Colonoscopy2014