DDA: Dimensionality Driven Augmentation Search for Contrastive Learning in Laparoscopic Surgery

TL;DR

This paper introduces DDA, a method that automatically searches for optimal data augmentation policies in contrastive learning for laparoscopic surgery, improving performance and providing domain-specific insights.

Contribution

DDA is a novel differentiable search method that optimizes augmentation policies based on local representation dimensionality for medical imaging applications.

Findings

DDA outperforms existing baselines in laparoscopic image classification and segmentation.

It identifies domain-specific augmentation strategies, such as the limited effectiveness of hue adjustments.

DDA offers insights into contrastive learning dependencies in medical imaging.

Abstract

Self-supervised learning (SSL) has potential for effective representation learning in medical imaging, but the choice of data augmentation is critical and domain-specific. It remains uncertain if general augmentation policies suit surgical applications. In this work, we automate the search for suitable augmentation policies through a new method called Dimensionality Driven Augmentation Search (DDA). DDA leverages the local dimensionality of deep representations as a proxy target, and differentiably searches for suitable data augmentation policies in contrastive learning. We demonstrate the effectiveness and efficiency of DDA in navigating a large search space and successfully identifying an appropriate data augmentation policy for laparoscopic surgery. We systematically evaluate DDA across three laparoscopic image classification and segmentation tasks, where it significantly improves over existing baselines. Furthermore, DDA's optimised set of augmentations provides insight into domain-specific dependencies when applying contrastive learning in medical applications. For example, while hue is an effective augmentation for natural images, it is not advantageous for laparoscopic images.