mlsanityData to 7 Oct 2026

Unraveling Normal Anatomy via Fluid-Driven Anomaly Randomization

arXiv 2501.13370 ↗First posted 23 Jan 2025eess.IV cs.CVcited by 4

Not on the map. Why?

Abstract (arXiv)

Data-driven machine learning has made significant strides in medical image analysis. However, most existing methods are tailored to specific modalities and assume a particular resolution (often isotropic). This limits their generalizability in clinical settings, where variations in scan appearance arise from differences in sequence parameters, resolution, and orientation. Furthermore, most general-purpose models are designed for healthy subjects and suffer from performance degradation when pathology is present. We introduce UNA (Unraveling Normal Anatomy), the first modality-agnostic learning approach for normal brain anatomy reconstruction that can handle both healthy scans and cases with pathology. We propose a fluid-driven anomaly randomization method that generates an unlimited number of realistic pathology profiles on-the-fly. UNA is trained on a combination of synthetic and real data, and can be applied directly to real images with potential pathology without the need for fine-tuning. We demonstrate UNA's effectiveness in reconstructing healthy brain anatomy and showcase its direct application to anomaly detection, using both simulated and real images from 3D healthy and stroke datasets, including CT and MRI scans. By bridging the gap between healthy and diseased images, UNA enables the use of general-purpose models on diseased images, opening up new opportunities for large-scale analysis of uncurated clinical images in the presence of pathology. Code is available at https://github.com/peirong26/UNA.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 30 of its 55 references that are in the map. First posted on arXiv 23 Jan 2025.

Builds on

30 of 55 references.

References · 30

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  2. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,385
  3. 2016V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentationcited by 2,284
  4. 2022Large Language Models Encode Clinical Knowledgecited by 1,807
  5. 2017PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volumecited by 966
  6. 2018VoxelMorph: A Learning Framework for Deformable Medical Image Registrationcited by 657
  7. 2018Deep Visual Domain Adaptation: A Surveycited by 634
  8. 2016Efficient Multi-Scale 3D CNN with Fully Connected CRF for Accurate Brain Lesion Segmentationcited by 263
  9. 2021SynthSeg: Segmentation of brain MRI scans of any contrast and resolution without retrainingcited by 235
  10. 2018A Deep Learning Framework for Unsupervised Affine and Deformable Image Registrationcited by 160
  11. 2020Autoencoders for Unsupervised Anomaly Segmentation in Brain MR Images: A Comparative Studycited by 146
  12. 2020SynthMorph: learning contrast-invariant registration without acquired imagescited by 137
18 more
  1. 2022SynthStrip: Skull-Stripping for Any Brain Imagecited by 126
  2. 2017Quicksilver: Fast Predictive Image Registration - a Deep Learning Approachcited by 113
  3. 2022ISLES 2022: A multi-center magnetic resonance imaging stroke lesion segmentation datasetcited by 104
  4. 2017Deep Learning for Physical Processes: Incorporating Prior Scientific Knowledgecited by 95
  5. 2020Local Temperature Scaling for Probability Calibrationcited by 95
  6. 2022Robust machine learning segmentation for large-scale analysis of heterogeneous clinical brain MRI datasetscited by 50
  7. 2020Joint super-resolution and synthesis of 1 mm isotropic MP-RAGE volumes from clinical MRI exams with scans of different orientation, resolution and contrastcited by 45
  8. 2021Accurate Point Cloud Registration with Robust Optimal Transportcited by 38
  9. 2024Learning General-Purpose Biomedical Volume Representations using Randomized Synthesiscited by 27
  10. 2023Unsupervised 3D out-of-distribution detection with latent diffusion modelscited by 22
  11. 2023Quantifying white matter hyperintensity and brain volumes in heterogeneous clinical and low-field portable MRIcited by 15
  12. 2024SAM-Med3D-MoE: Towards a Non-Forgetting Segment Anything Model via Mixture of Experts for 3D Medical Image Segmentationcited by 12
  13. 2023Brain-ID: Learning Contrast-agnostic Anatomical Representations for Brain Imagingcited by 10
  14. 2020Discovering Hidden Physics Behind Transport Dynamicscited by 8
  15. 2024PEPSI: Pathology-Enhanced Pulse-Sequence-Invariant Representations for Brain MRIcited by 8
  16. 2023HelmFluid: Learning Helmholtz Dynamics for Interpretable Fluid Predictioncited by 7
  17. 2023Learning to Estimate Single-View Volumetric Flow Motions without 3D Supervisioncited by 5
  18. 2022Fluid registration between lung CT and stationary chest tomosynthesis imagescited by 4

Built on by

Cited by year

Citing papers per year: 2025: 4.

4 papers.

  1. 2025Conditional diffusion models for guided anomaly detection in brain images using fluid-driven anomaly randomizationcited by 6
  2. 2025USB: Unified Synthetic Brain Framework for Bidirectional Pathology-Healthy Generation and Editingcited by 1
  3. 2025Enhancing Corpus Callosum Segmentation in Fetal MRI via Pathology-Informed Domain Randomizationcited by 0
  4. 2025Generating healthy counterfactuals with denoising diffusion bridge modelscited by 0