mlsanityData to 14 Oct 2026

Leveraging Spatial Context for Positive Pair Sampling in Histopathology Image Representation Learning

arXiv 2503.05170 ↗First posted 7 Mar 2025cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Deep learning has shown strong potential in cancer classification from whole-slide images (WSIs), but the need for extensive expert annotations often limits its success. Annotation-free approaches, such as multiple instance learning (MIL) and self-supervised learning (SSL), have emerged as promising alternatives to traditional annotation-based methods. However, conventional SSL methods typically rely on synthetic data augmentations, which may fail to capture the spatial structure critical to histopathology. In this work, we propose a spatial context-driven positive pair sampling strategy that enhances SSL by leveraging the morphological coherence of spatially adjacent patches within WSIs. Our method is modular and compatible with established joint embedding SSL frameworks, including Barlow Twins, BYOL, VICReg, and DINOv2. We evaluate its effectiveness on both slide-level classification using MIL and patch-level linear probing. Experiments across four datasets demonstrate consistent performance improvements, with accuracy gains of 5% to 10% compared to standard augmentation-based sampling. These findings highlight the value of spatial context in improving representation learning for computational pathology and provide a biologically meaningful enhancement for pretraining models in annotation-limited settings. The code is available at https://anonymous.4open.science/r/contextual-pairs-E72F/.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 31 of its 50 references that are in the map. First posted on arXiv 7 Mar 2025.

Builds on

31 of 50 references.

References · 31

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  3. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  4. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  5. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  6. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,750
  7. 2014Going Deeper with Convolutionscited by 4,241
  8. 2019RandAugment: Practical automated data augmentation with a reduced search spacecited by 2,269
  9. 2021Barlow Twins: Self-Supervised Learning via Redundancy Reductioncited by 1,838
  10. 2018Attention-based Deep Multiple Instance Learningcited by 1,095
  11. 2017Large Batch Training of Convolutional Networkscited by 535
  12. 2020Dual-stream Multiple Instance Learning Network for Whole Slide Image Classification with Self-supervised Contrastive Learningcited by 478
19 more
  1. 2022DTFD-MIL: Double-Tier Feature Distillation Multiple Instance Learning for Histopathology Whole Slide Image Classificationcited by 256
  2. 2022Benchmarking Self-Supervised Learning on Diverse Pathology Datasetscited by 150
  3. 2018Medical Image Synthesis for Data Augmentation and Anonymization using Generative Adversarial Networkscited by 147
  4. 2018Tile2Vec: Unsupervised representation learning for spatially distributed datacited by 110
  5. 2020Self-Path: Self-supervision for Classification of Pathology Images with Limited Annotationscited by 89
  6. 2021GANs for Medical Image Synthesis: An Empirical Studycited by 65
  7. 2024Hibou: A Family of Foundational Vision Transformers for Pathologycited by 57
  8. 2020A survey on Semi-, Self- and Unsupervised Learning for Image Classificationcited by 55
  9. 2017Constrained Deep Weak Supervision for Histopathology Image Segmentationcited by 51
  10. 2019Context-Aware Convolutional Neural Network for Grading of Colorectal Cancer Histology Imagescited by 39
  11. 2018Classification and Disease Localization in Histopathology Using Only Global Labels: A Weakly-Supervised Approachcited by 33
  12. 2022Transformer based multiple instance learning for weakly supervised histopathology image segmentationcited by 26
  13. 2021Focus on the Positives: Self-Supervised Learning for Biodiversity Monitoringcited by 16
  14. 2019Automated Classification of Histopathology Images Using Transfer Learningcited by 14
  15. 2021SimTriplet: Simple Triplet Representation Learning with a Single GPUcited by 12
  16. 2023Improving Representation Learning for Histopathologic Images with Cluster Constraintscited by 8
  17. 2022Kernel Attention Transformer (KAT) for Histopathology Whole Slide Image Classificationcited by 7
  18. 2021HistoTransfer: Understanding Transfer Learning for Histopathologycited by 6
  19. 2023Histopathology Image Classification using Deep Manifold Contrastive Learningcited by 3

Built on by

Not cited yet.