Improvement Strategies for Few-Shot Learning in OCT Image Classification of Rare Retinal Diseases

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tai, Cheng-Yu, Chen, Ching-Wen, Wu, Chi-Chin, Chiu, Bo-Chen, Cheng-Hung, Lin, Lu, Cheng-Kai, Wang, Jia-Kang, Huang, Tzu-Lun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912395383799808
author Tai, Cheng-Yu
Chen, Ching-Wen
Wu, Chi-Chin
Chiu, Bo-Chen
Cheng-Hung
Lin
Lu, Cheng-Kai
Wang, Jia-Kang
Huang, Tzu-Lun
author_facet Tai, Cheng-Yu
Chen, Ching-Wen
Wu, Chi-Chin
Chiu, Bo-Chen
Cheng-Hung
Lin
Lu, Cheng-Kai
Wang, Jia-Kang
Huang, Tzu-Lun
contents This paper focuses on using few-shot learning to improve the accuracy of classifying OCT diagnosis images with major and rare classes. We used the GAN-based augmentation strategy as a baseline and introduced several novel methods to further enhance our model. The proposed strategy contains U-GAT-IT for improving the generative part and uses the data balance technique to narrow down the skew of accuracy between all categories. The best model obtained was built with CBAM attention mechanism and fine-tuned InceptionV3, and achieved an overall accuracy of 97.85%, representing a significant improvement over the original baseline.
format Preprint
id arxiv_https___arxiv_org_abs_2505_20149
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Improvement Strategies for Few-Shot Learning in OCT Image Classification of Rare Retinal Diseases
Tai, Cheng-Yu
Chen, Ching-Wen
Wu, Chi-Chin
Chiu, Bo-Chen
Cheng-Hung
Lin
Lu, Cheng-Kai
Wang, Jia-Kang
Huang, Tzu-Lun
Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
This paper focuses on using few-shot learning to improve the accuracy of classifying OCT diagnosis images with major and rare classes. We used the GAN-based augmentation strategy as a baseline and introduced several novel methods to further enhance our model. The proposed strategy contains U-GAT-IT for improving the generative part and uses the data balance technique to narrow down the skew of accuracy between all categories. The best model obtained was built with CBAM attention mechanism and fine-tuned InceptionV3, and achieved an overall accuracy of 97.85%, representing a significant improvement over the original baseline.
title Improvement Strategies for Few-Shot Learning in OCT Image Classification of Rare Retinal Diseases
topic Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.20149