Keep Your Friends Close & Enemies Farther: Debiasing Contrastive Learning with Spatial Priors in 3D Radiology Images

TL;DR

This paper introduces Spade, a 3D contrastive learning framework that uses spatial correspondences in radiology images to improve segmentation by effectively selecting positive and negative samples, enhancing representation learning.

Contribution

The paper presents a novel contrastive learning approach leveraging spatial correspondences to improve 3D radiology image segmentation, with tailored selection strategies for global and local features.

Findings

Spade outperforms recent methods on three segmentation tasks.

Utilizes spatial correspondences without annotations or high computational costs.

Learns both invariant and equivariant representations for better segmentation.

Abstract

Understanding of spatial attributes is central to effective 3D radiology image analysis where crop-based learning is the de facto standard. Given an image patch, its core spatial properties (e.g., position & orientation) provide helpful priors on expected object sizes, appearances, and structures through inherent anatomical consistencies. Spatial correspondences, in particular, can effectively gauge semantic similarities between inter-image regions, while their approximate extraction requires no annotations or overbearing computational costs. However, recent 3D contrastive learning approaches either neglect correspondences or fail to maximally capitalize on them. To this end, we propose an extensible 3D contrastive framework (Spade, for Spatial Debiasing) that leverages extracted correspondences to select more effective positive & negative samples for representation learning. Our method learns both globally invariant and locally equivariant representations with downstream segmentation in mind. We also propose separate selection strategies for global & local scopes that tailor to their respective representational requirements. Compared to recent state-of-the-art approaches, Spade shows notable improvements on three downstream segmentation tasks (CT Abdominal Organ, CT Heart, MR Heart).