mlsanityData to 14 Oct 2026

FastSurferVINN: Building Resolution-Independence into Deep Learning Segmentation Methods -- A Solution for HighRes Brain MRI

arXiv 2112.09654 ↗First posted 17 Dec 2021eess.IV cs.CVcited by 19

Not on the map. Why?

Abstract (arXiv)

Leading neuroimaging studies have pushed 3T MRI acquisition resolutions below 1.0 mm for improved structure definition and morphometry. Yet, only few, time-intensive automated image analysis pipelines have been validated for high-resolution (HiRes) settings. Efficient deep learning approaches, on the other hand, rarely support more than one fixed resolution (usually 1.0 mm). Furthermore, the lack of a standard submillimeter resolution as well as limited availability of diverse HiRes data with sufficient coverage of scanner, age, diseases, or genetic variance poses additional, unsolved challenges for training HiRes networks. Incorporating resolution-independence into deep learning-based segmentation, i.e., the ability to segment images at their native resolution across a range of different voxel sizes, promises to overcome these challenges, yet no such approach currently exists. We now fill this gap by introducing a Voxelsize Independent Neural Network (VINN) for resolution-independent segmentation tasks and present FastSurferVINN, which (i) establishes and implements resolution-independence for deep learning as the first method simultaneously supporting 0.7-1.0 mm whole brain segmentation, (ii) significantly outperforms state-of-the-art methods across resolutions, and (iii) mitigates the data imbalance problem present in HiRes datasets. Overall, internal resolution-independence mutually benefits both HiRes and 1.0 mm MRI segmentation. With our rigorously validated FastSurferVINN we distribute a rapid tool for morphometric neuroimage analysis. The VINN architecture, furthermore, represents an efficient resolution-independent segmentation method for wider application

Key facts. Cited by 19 papers in the map (data through 14 Oct 2026). Builds on 34 of its 122 references that are in the map. First posted on arXiv 17 Dec 2021.

Builds on

34 of 122 references.

References · 34

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  2. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 5,045
  3. 2015Spatial Transformer Networkscited by 1,429
  4. 2018Moment Matching for Multi-Source Domain Adaptationcited by 1,331
  5. 2014Image Super-Resolution Using Deep Convolutional Networkscited by 1,224
  6. 2017CyCADA: Cycle-Consistent Adversarial Domain Adaptationcited by 1,082
  7. 2015Accurate Image Super-Resolution Using Very Deep Convolutional Networkscited by 903
  8. 2015Deeply-Recursive Convolutional Network for Image Super-Resolutioncited by 317
  9. 2016Efficient Multi-Scale 3D CNN with Fully Connected CRF for Accurate Brain Lesion Segmentationcited by 263
  10. 2013Maxout Networkscited by 187
  11. 2015Attention to Scale: Scale-aware Semantic Image Segmentationcited by 169
  12. 2019Scale-Aware Trident Networks for Object Detectioncited by 159
22 more
  1. 2016The One Hundred Layers Tiramisu: Fully Convolutional DenseNets for Semantic Segmentationcited by 122
  2. 2019FastSurfer -- A fast and accurate deep learning based neuroimaging pipelinecited by 112
  3. 2018Deep Cocktail Network: Multi-source Unsupervised Domain Adaptation with Category Shiftcited by 109
  4. 2019Multi-source Domain Adaptation for Semantic Segmentationcited by 74
  5. 2018Joint Optic Disc and Cup Segmentation Based on Multi-label Deep Network and Polar Transformationcited by 73
  6. 2018Unsupervised Multi-Target Domain Adaptation: An Information Theoretic Approachcited by 63
  7. 2018Multi-Source Domain Adaptation with Mixture of Expertscited by 62
  8. 20193D Whole Brain Segmentation using Spatially Localized Atlas Network Tilescited by 50
  9. 2020HookNet: multi-resolution convolutional neural networks for semantic segmentation in histopathology whole-slide imagescited by 45
  10. 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
  11. 2018QuickNAT: A Fully Convolutional Network for Quick and Accurate Segmentation of Neuroanatomycited by 37
  12. 2019Blending-target Domain Adaptation by Adversarial Meta-Adaptation Networkscited by 37
  13. 2021Curriculum Graph Co-Teaching for Multi-Target Domain Adaptationcited by 37
  14. 2017DeepNAT: Deep Convolutional Neural Network for Segmenting Neuroanatomycited by 25
  15. 2019AssemblyNet: A large ensemble of CNNs for 3D Whole Brain MRI Segmentationcited by 21
  16. 2020Partial Volume Segmentation of Brain MRI Scans of any Resolution and Contrastcited by 16
  17. 2018Multi-Resolution Networks for Semantic Segmentation in Whole Slide Imagescited by 14
  18. 2018Autofocus Layer for Semantic Segmentationcited by 13
  19. 2018Knowing what you know in brain segmentation using Bayesian deep neural networkscited by 13
  20. 2017Error Corrective Boosting for Learning Fully Convolutional Networks with Limited Datacited by 12
  21. 2018Attention to Refine through Multi-Scales for Semantic Segmentationcited by 5
  22. 2020ASDN: A Deep Convolutional Network for Arbitrary Scale Image Super-Resolutioncited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 1; 2023: 4; 2024: 4; 2025: 6; 2026: 4.

19 papers.

Sort:

By area:

  1. 2023Low-field magnetic resonance image enhancement via stochastic image quality transfercited by 19
  2. 2025Regression is all you need for medical image translationcited by 8
  3. 2025A Modality-agnostic Multi-task Foundation Model for Human Brain Imagingcited by 8
  4. 2024Recon-all-clinical: Cortical surface reconstruction and analysis of heterogeneous clinical brain MRIcited by 7
  5. 2023FastSurfer-HypVINN: Automated sub-segmentation of the hypothalamus and adjacent structures on high-resolutional brain MRIcited by 2
  6. 2023TABSurfer: a Hybrid Deep Learning Architecture for Subcortical Segmentationcited by 2
  7. 2024MedSegMamba: 3D CNN-Mamba Hybrid Architecture for Brain Segmentationcited by 2
  8. 2025From Low Field to High Value: Robust Cortical Mapping from Low-Field MRIcited by 2
  9. 2026SIAM: Head and Brain MRI Segmentation from Few High-Quality Templates via Synthetic Trainingcited by 2
  10. 2023VINNA for Neonates -- Orientation Independence through Latent Augmentationscited by 1
  11. 2024Multi-Domain Data Aggregation for Axon and Myelin Segmentation in Histology Imagescited by 1
  12. 2025UNSURF: Uncertainty Quantification for Cortical Surface Reconstruction of Clinical Brain MRIscited by 1
7 more
  1. 2022Identifying and Combating Bias in Segmentation Networks by leveraging multiple resolutionscited by 0
  2. 2024Harnessing The Power of Attention For Patch-Based Biomedical Image Classificationcited by 0
  3. 2025Uncertain but Useful: Leveraging CNN Training Variability into Data Augmentationcited by 0
  4. 2025FastSurfer-CC: A robust, accurate, and comprehensive framework for corpus callosum morphometrycited by 0
  5. 2026GOUHFI 2.0: A Next-Generation Toolbox for Brain Segmentation and Cortex Parcellation at Ultra-High Field MRIcited by 0
  6. 2026Native space based pipelines outperform template space based pipeline in subcortical segmentationcited by 0
  7. 2026Stride Independent Patching for Deep Learningcited by 0