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 Concept Videos

Brain Imaging01:14

Brain Imaging

824
Brain imaging technologies provide critical insights into both the structure and function of the human brain, enabling medical professionals and researchers to diagnose, study, and treat neurological disorders or psychiatric disorders more effectively.
These technologies include computerized axial tomography (CAT or CT scans), positron-emission tomography (PET scans),  magnetic resonance imaging (MRI),  functional magnetic resonance imaging (fMRI), and Transcranial Magnetic...
824

You might also read

Related Articles

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

Sort by
Same author

The seeds of the future are in the present: A blind exploration of metastable states.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same author

Cardiometabolic health and physical robustness map onto distinct patterns of brain structure and neurotransmitter systems.

PLoS biology·2025
Same author

Schedule-Robust Continual Learning.

IEEE transactions on pattern analysis and machine intelligence·2025
Same author

Fast and Fourier features for transfer learning of interatomic potentials.

npj computational materials·2025
Same author

Slow dynamical modes from static averages.

The Journal of chemical physics·2025
Same author

Voxelwise Multivariate Analysis of Brain-Psychosocial Associations in Adolescents Reveals 6 Latent Dimensions of Cognition and Psychopathology.

Biological psychiatry. Cognitive neuroscience and neuroimaging·2024

Related Experiment Video

Updated: Mar 6, 2026

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks
11:18

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks

Published on: March 2, 2015

10.9K

Sparsity Is Better with Stability: Combining Accuracy and Stability for Model Selection in Brain Decoding.

Luca Baldassarre1, Massimiliano Pontil2, Janaina Mourão-Miranda3

  • 1Laboratory for Information and Inference Systems, École Polytechnique Fédérale de Lausanne (EPFL) Lausanne, Switzerland.

Frontiers in Neuroscience
|March 7, 2017
PubMed
Summary
This summary is machine-generated.

Structured sparse methods in neuroimaging can be unstable. Adding stability criteria, like solution overlap, improves model reproducibility and mitigates instability in brain decoding with functional magnetic resonance imaging (fMRI).

Keywords:
model selectionpredictive modelsreproducibilitysparse methodsstructured sparsity

More Related Videos

P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation
06:09

P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation

Published on: September 8, 2023

996
Analyzing Neural Activity and Connectivity Using Intracranial EEG Data with SPM Software
06:50

Analyzing Neural Activity and Connectivity Using Intracranial EEG Data with SPM Software

Published on: October 30, 2018

10.0K

Related Experiment Videos

Last Updated: Mar 6, 2026

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks
11:18

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks

Published on: March 2, 2015

10.9K
P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation
06:09

P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation

Published on: September 8, 2023

996
Analyzing Neural Activity and Connectivity Using Intracranial EEG Data with SPM Software
06:50

Analyzing Neural Activity and Connectivity Using Intracranial EEG Data with SPM Software

Published on: October 30, 2018

10.0K

Area of Science:

  • Neuroimaging
  • Machine Learning
  • Computational Neuroscience

Background:

  • Structured sparse methods are increasingly used in neuroimaging for their interpretability.
  • Sparsity, while promoting interpretability, can lead to unstable predictive models, particularly in functional magnetic resonance imaging (fMRI) brain decoding.
  • Existing methods like LASSO and Elastic Net lack robustness to variations in data or experimental conditions.

Purpose of the Study:

  • To investigate the impact of stability/reproducibility as a model selection criterion in sparse and structured sparse methods for fMRI brain decoding.
  • To compare the effectiveness of different model selection criteria, including classification accuracy alone, accuracy combined with solution overlap, and accuracy combined with solution correlation.
  • To evaluate the stability and interpretability of various sparse methods under different selection criteria.

Main Methods:

  • Applied several sparse and structured sparse methods including LASSO, Elastic Net, Total Variation, sparse Total Variation, Laplacian, and Graph Laplacian Elastic Net (GraphNET).
  • Compared three model selection criteria: classification accuracy only, classification accuracy plus solution overlap, and classification accuracy plus solution correlation.
  • Assessed model stability and reproducibility by examining the similarity of solutions across different conditions or subsamples.

Main Results:

  • Explicitly incorporating stability/reproducibility into model optimization mitigates the inherent instability of sparse methods.
  • Using classification accuracy and overlap between solutions as a joint criterion leads to more consistent results across different sparsity methods.
  • This joint criterion improves similarity in accuracy, sparsity levels, and coefficient maps compared to using accuracy alone.

Conclusions:

  • Stability/reproducibility should be considered alongside accuracy for robust model selection in neuroimaging.
  • The combination of accuracy and overlap offers a promising approach to enhance the reliability of sparse methods in fMRI brain decoding.
  • This strategy can lead to more stable and reproducible neuroimaging analysis, advancing the field of brain decoding.