Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Mindboggle: a scatterbrained approach to automate brain labeling.

Arno Klein1, Joy Hirsch

  • 1fMRI Research Center, Columbia University, New York 10032, USA. arno@binarybottle.com

Neuroimage
|January 4, 2005
PubMed
Summary

Mindboggle, an automated human brain MRI labeling tool, outperforms other methods by disassembling and reassembling an atlas to match subject brains. It demonstrates superior accuracy and robustness, even with incomplete or artifact-affected data.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

An open, fully-processed data resource for studying mood and sleep variability in the developing brain.

Aperture neuro·2026
Same author

Listening to a Consonant Chord Progression during Live Face-to-Face Gaze Enhances Neural Activity in Social Systems.

The Journal of neuroscience : the official journal of the Society for Neuroscience·2026
Same author

Patterns of ongoing thought in the real world and their links to mental health and well-being.

PLOS mental health·2026
Same author

Investigating face processing in online interactions via UK-US hyperscanning using fNIRS.

Imaging neuroscience (Cambridge, Mass.)·2026
Same author

Accurate and efficient data-driven psychiatric assessment using machine learning.

BMC medical informatics and decision making·2026
Same author

Real-time Ecological Assessment of the Context of mental and physical Health (REACH) on Curious.

Research square·2025

Area of Science:

  • Neuroimaging
  • Computational Neuroscience
  • Medical Image Analysis

Background:

  • Accurate anatomical labeling of human brain MRI data is crucial for neuroscience research.
  • Existing automated methods often struggle with inter-individual variability and data quality issues.

Purpose of the Study:

  • To introduce and evaluate Mindboggle, a novel automated approach for anatomical labeling of human brain MRI.
  • To compare Mindboggle's performance against established linear and nonlinear registration methods.

Main Methods:

  • Mindboggle employs a feature-matching strategy, disassembling and reassembling a labeled atlas to match subject brains.
  • The process involves converting MRI data into sulcus pieces, matching atlas to subject pieces, transforming boundaries, warping labels, and propagating labels.

Related Experiment Videos

  • Comparison with linear registration, SPM2, AIR, and ANIMAL was conducted using voxel-wise agreement metrics.

    • Main Results:

    • Mindboggle significantly outperformed linear registration, SPM2, AIR, and ANIMAL in automated brain MRI labeling accuracy.
    • Its performance remained superior even after applying post-processing label-filling steps to other methods.
    • Mindboggle demonstrated high robustness against image artifacts, poor quality, and incomplete brain data, with performance not degrading significantly with simulated lesions.

    Conclusions:

    • Mindboggle offers a highly accurate and robust automated solution for anatomical labeling of human brain MRI.
    • Its unique feature-matching and reassembly approach overcomes limitations of traditional registration methods.
    • Mindboggle represents a significant advancement for large-scale neuroimaging studies requiring reliable brain structure identification.