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

arXiv: https://arxiv.org/abs/2211.08643 · HTML: https://mlsanity.com/paper/2211.08643/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 26 references that are in the map.
- First posted on arXiv 16 Nov 2022.

## Builds on (most cited first, 21 of 21)

- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Improved Baselines with Momentum Contrastive Learning](https://mlsanity.com/paper/2003.04297/) (2020)
- [Unsupervised Feature Learning via Non-Parametric Instance-level Discrimination](https://mlsanity.com/paper/1805.01978/) (2018)
- [Res2Net: A New Multi-scale Backbone Architecture](https://mlsanity.com/paper/1904.01169/) (2019)
- [Debiased Contrastive Learning](https://mlsanity.com/paper/2007.00224/) (2020)
- [AMOS: A Large-Scale Abdominal Multi-Organ Benchmark for Versatile Medical Image Segmentation](https://mlsanity.com/paper/2206.08023/) (2022)
- [Big Self-Supervised Models Advance Medical Image Classification](https://mlsanity.com/paper/2101.05224/) (2021)
- [Contrastive learning of global and local features for medical image segmentation with limited annotations](https://mlsanity.com/paper/2006.10511/) (2020)
- [Models Genesis: Generic Autodidactic Models for 3D Medical Image Analysis](https://mlsanity.com/paper/1908.06912/) (2019)
- [Boosting Contrastive Self-Supervised Learning with False Negative Cancellation](https://mlsanity.com/paper/2011.11765/) (2020)
- [SimCVD: Simple Contrastive Voxel-Wise Representation Distillation for Semi-Supervised Medical Image Segmentation](https://mlsanity.com/paper/2108.06227/) (2021)
- [Transferable Visual Words: Exploiting the Semantics of Anatomical Patterns for Self-supervised Learning](https://mlsanity.com/paper/2102.10680/) (2021)
- [MoCo-CXR: MoCo Pretraining Improves Representation and Transferability of Chest X-ray Models](https://mlsanity.com/paper/2010.05352/) (2020)
- [Preservational Learning Improves Self-supervised Medical Image Models by Reconstructing Diverse Contexts](https://mlsanity.com/paper/2109.04379/) (2021)
- [Positional Contrastive Learning for Volumetric Medical Image Segmentation](https://mlsanity.com/paper/2106.09157/) (2021)
- [Revisiting Rubik's Cube: Self-supervised Learning with Volume-wise Transformation for 3D Medical Image Segmentation](https://mlsanity.com/paper/2007.08826/) (2020)
- [PGL: Prior-Guided Local Self-supervised Learning for 3D Medical Image Segmentation](https://mlsanity.com/paper/2011.12640/) (2020)
- [Hierarchical Self-Supervised Learning for Medical Image Segmentation Based on Multi-Domain Data Aggregation](https://mlsanity.com/paper/2107.04886/) (2021)
- [Towards Fine-grained Visual Representations by Combining Contrastive Learning with Image Reconstruction and Attention-weighted Pooling](https://mlsanity.com/paper/2104.04323/) (2021)
- [SAR: Scale-Aware Restoration Learning for 3D Tumor Segmentation](https://mlsanity.com/paper/2010.06107/) (2020)

## Built on by (most cited first, 5 of 5)

- [I-MedSAM: Implicit Medical Image Segmentation with Segment Anything](https://mlsanity.com/paper/2311.17081/) (2023)
- [Self Pre-training with Topology- and Spatiality-aware Masked Autoencoders for 3D Medical Image Segmentation](https://mlsanity.com/paper/2406.10519/) (2024)
- [Sli2Vol+: Segmenting 3D Medical Images Based on an Object Estimation Guided Correspondence Flow Network](https://mlsanity.com/paper/2411.13873/) (2024)
- [SwIPE: Efficient and Robust Medical Image Segmentation with Implicit Patch Embeddings](https://mlsanity.com/paper/2307.12429/) (2023)
- [Patch-MoE Mamba: A Patch-Ordered Mixture-of-Experts State Space Architecture for Medical Image Segmentation](https://mlsanity.com/paper/2605.17719/) (2026)

## Abstract (arXiv)

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).