ACE: Anatomically Consistent Embeddings in Composition and Decomposition
Not on the map. Why?
Abstract (arXiv)
Medical images acquired from standardized protocols show consistent macroscopic or microscopic anatomical structures, and these structures consist of composable/decomposable organs and tissues, but existing self-supervised learning (SSL) methods do not appreciate such composable/decomposable structure attributes inherent to medical images. To overcome this limitation, this paper introduces a novel SSL approach called ACE to learn anatomically consistent embedding via composition and decomposition with two key branches: (1) global consistency, capturing discriminative macro-structures via extracting global features; (2) local consistency, learning fine-grained anatomical details from composable/decomposable patch features via corresponding matrix matching. Experimental results across 6 datasets 2 backbones, evaluated in few-shot learning, fine-tuning, and property analysis, show ACE's superior robustness, transferability, and clinical potential. The innovations of our ACE lie in grid-wise image cropping, leveraging the intrinsic properties of compositionality and decompositionality of medical images, bridging the semantic gap from high-level pathologies to low-level tissue anomalies, and providing a new SSL method for medical imaging.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 47 of its 67 references that are in the map. First posted on arXiv 17 Jan 2025.
Builds on
47 of 67 references.
References · 47
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
- 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
- 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
- 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
- 2015Distilling the Knowledge in a Neural Networkcited by 10,324
- 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
- 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
- 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,733
- 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,750
- 2020Exploring Simple Siamese Representation Learningcited by 2,866
- 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,857
35 more
- 2021Barlow Twins: Self-Supervised Learning via Redundancy Reductioncited by 1,838
- 2019Deep High-Resolution Representation Learning for Human Pose Estimationcited by 1,339
- 2019Contrastive Multiview Codingcited by 1,301
- 2018Unified Perceptual Parsing for Scene Understandingcited by 1,271
- 2017ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseasescited by 972
- 2019Self-supervised Visual Feature Learning with Deep Neural Networks: A Surveycited by 657
- 2020Dense Contrastive Learning for Self-Supervised Visual Pre-Trainingcited by 512
- 2020Contrastive learning of global and local features for medical image segmentation with limited annotationscited by 309
- 2020Propagate Yourself: Exploring Pixel-Level Consistency for Unsupervised Visual Representation Learningcited by 297
- 2019Models Genesis: Generic Autodidactic Models for 3D Medical Image Analysiscited by 159
- 2023Knowledge-enhanced Visual-Language Pre-training on Chest Radiology Imagescited by 128
- 2021How to represent part-whole hierarchies in a neural networkcited by 123
- 2022VICRegL: Self-Supervised Learning of Local Visual Featurescited by 118
- 2021Region Similarity Representation Learningcited by 96
- 2021Transferable Visual Words: Exploiting the Semantics of Anatomical Patterns for Self-supervised Learningcited by 68
- 2020SAM: Self-supervised Learning of Pixel-wise Anatomical Embeddings in Radiological Imagescited by 67
- 2022Patch-level Representation Learning for Self-supervised Vision Transformerscited by 60
- 2023LVM-Med: Learning Large-Scale Self-Supervised Vision Models for Medical Imaging via Second-order Graph Matchingcited by 60
- 2019Multimodal Self-Supervised Learning for Medical Image Analysiscited by 57
- 2022DiRA: Discriminative, Restorative, and Adversarial Learning for Self-supervised Medical Image Analysiscited by 48
- 2021Compressive Visual Representationscited by 41
- 2020ChestX-Det10: Chest X-ray Dataset on Detection of Thoracic Abnormalitiescited by 38
- 2020Learning Semantics-enriched Representation via Self-discovery, Self-classification, and Self-restorationcited by 35
- 2021A Structure-Aware Relation Network for Thoracic Diseases Detection and Segmentationcited by 34
- 2022Leverage Your Local and Global Representations: A New Self-Supervised Learning Strategycited by 30
- 2023Anatomical Invariance Modeling and Semantic Alignment for Self-supervised Learning in 3D Medical Image Analysiscited by 26
- 2023DropPos: Pre-Training Vision Transformers by Reconstructing Dropped Positionscited by 19
- 2024Representing Part-Whole Hierarchies in Foundation Models by Learning Localizability, Composability, and Decomposability from Anatomy via Self-Supervisioncited by 13
- 2023Towards Foundation Models Learned from Anatomy in Medical Imaging via Self-Supervisioncited by 11
- 2023Foundation Ark: Accruing and Reusing Knowledge for Superior and Robust Performancecited by 11
- 2024DeViDe: Faceted medical knowledge for improved medical vision-language pre-trainingcited by 10
- 2022Anatomy-Aware Contrastive Representation Learning for Fetal Ultrasoundcited by 8
- 2023Learning Anatomically Consistent Embedding for Chest Radiographycited by 8
- 2023DenseDINO: Boosting Dense Self-Supervised Learning with Token-Based Point-Level Consistencycited by 6
- 2022Precise Location Matching Improves Dense Contrastive Learning in Digital Pathologycited by 4
Built on by
Citing papers per year: 2026: 1.
1 paper.