Comparative Analysis of Data Augmentation for Retinal OCT Biomarker Segmentation

TL;DR

This study systematically evaluates how different data augmentation techniques affect retinal OCT biomarker segmentation, emphasizing that their effectiveness depends on dataset size and characteristics, and advocating for a strategic, context-aware approach.

Contribution

It provides an exhaustive analysis of various data augmentation methods for retinal OCT segmentation, highlighting the importance of dataset-specific strategies and revealing their variable effectiveness.

Findings

Augmentation benefits are greater with limited labeled data.

Transformation-based methods are particularly effective in scarce data scenarios.

Effectiveness of augmentation varies with dataset characteristics.

Abstract

Data augmentation plays a crucial role in addressing the challenge of limited expert-annotated datasets in deep learning applications for retinal Optical Coherence Tomography (OCT) scans. This work exhaustively investigates the impact of various data augmentation techniques on retinal layer boundary and fluid segmentation. Our results reveal that their effectiveness significantly varies based on the dataset's characteristics and the amount of available labeled data. While the benefits of augmentation are not uniform - being more pronounced in scenarios with scarce data, particularly for transformation-based methods - the findings highlight the necessity of a strategic approach to data augmentation. It is essential to note that the effectiveness of data augmentation varies significantly depending on the characteristics of the dataset. The findings emphasize the need for a nuanced approach, considering factors like dataset characteristics, the amount of labelled data, and the choice of model architecture.