Stronger is not better: Better Augmentations in Contrastive Learning for Medical Image Segmentation

TL;DR

This paper investigates the impact of various data augmentation strategies on contrastive learning for medical image segmentation, revealing that stronger augmentations do not always lead to better performance and proposing alternative augmentations that improve results.

Contribution

The study evaluates the effectiveness of strong data augmentations in contrastive learning for medical image segmentation and introduces alternative augmentation techniques that enhance performance.

Findings

Existing augmentations do not always improve segmentation performance.

Alternative augmentations can lead to better results.

Strong augmentations may sometimes degrade performance.

Abstract

Self-supervised contrastive learning is among the recent representation learning methods that have shown performance gains in several downstream tasks including semantic segmentation. This paper evaluates strong data augmentation, one of the most important components for self-supervised contrastive learning's improved performance. Strong data augmentation involves applying the composition of multiple augmentation techniques on images. Surprisingly, we find that the existing data augmentations do not always improve performance for semantic segmentation for medical images. We experiment with other augmentations that provide improved performance.