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

TL;DR

This paper presents novel strategies to enhance few-shot learning for OCT image classification of rare retinal diseases, achieving high accuracy through advanced augmentation and attention mechanisms.

Contribution

It introduces a combined approach using U-GAT-IT, data balancing, CBAM attention, and fine-tuned InceptionV3 to significantly improve classification accuracy in few-shot learning scenarios.

Findings

Achieved 97.85% overall accuracy.

Enhanced generative quality with U-GAT-IT.

Balanced data improves model performance.

Abstract

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.