Related Experiment Video
Updated: May 31, 2026
09:06
Whole-brain Segmentation and Change-point Analysis of Anatomical Brain MRI—Application in Premanifest Huntington's Disease
Published on: June 9, 2018
Fast segmentation with the NextBrain histological atlas
Oula Puonti1,2, Jackson Nolan2, Robert Dicamillo2
1Danish Research Centre for Magnetic Resonance, Centre for Functional and Diagnostic Imaging and Research, Copenhagen University Hospital - Amager and Hvidovre, Copenhagen, Denmark.
Imaging Neuroscience (Cambridge, Mass.)
|May 29, 2026
Summary
A new open-source tool significantly accelerates brain subregion segmentation for neuroimaging studies. This method provides accurate, high-resolution anatomical analysis at a fraction of the original computational cost.
Area of Science:
- Neuroimaging
- Computational Neuroscience
- Medical Image Analysis
Background:
- Subregion brain analysis is crucial for understanding aging and neurodegenerative diseases.
- The NextBrain histological atlas aids fine-grained investigations but has computationally intensive segmentation.
- Existing methods are prohibitive for large-scale neuroimaging studies.
Purpose of the Study:
- To develop an open-source tool for rapid, accurate brain subregion segmentation.
- To overcome the computational limitations of previous segmentation frameworks.
- To enable large-scale, high-resolution anatomical analysis.
Main Methods:
- A hybrid approach combining machine learning, contrast-adaptive segmentation, image synthesis, and diffeomorphic registration.
- GPU acceleration for all key computational steps.
- Validation across four modalities (in vivo MRI, ex vivo MRI, HiP-CT, photography) and ~4,000 scans.
Main Results:
- The accelerated tool achieves segmentation comparable in accuracy (Dice scores) to the original method.
- Runtime is reduced by over an order of magnitude (<5 minutes on GPU).
- The method is effective across diverse MRI resolutions and contrasts, including in vivo and ex vivo scans.
Conclusions:
- This work provides a practical, computationally efficient solution for high-resolution brain anatomical analysis at scale.
- The tool democratizes detailed neuroanatomical studies, facilitating research in aging and neurodegeneration.
- Public availability in FreeSurfer enhances accessibility for the neuroimaging community.
