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

You might also read

Related Articles

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

Sort by
Same author

Blood-Pressure Targets in Hypertension Management.

The New England journal of medicine·2026
Same author

Soda consumption and risk of dementia: The Northern Manhattan study.

Journal of Alzheimer's disease : JAD·2026
Same author

Biomarkers.

Alzheimer's & dementia : the journal of the Alzheimer's Association·2025
Same author

Your brain doesn't look a day past 70! Cross-sectional associations with brain-predicted age in the cognitively-intact oldest-old.

GeroScience·2025
Same author

Acute Care Utilization in Medicaid Enrollees Differs by Patient Rurality, Race, and Ethnicity.

Medical care·2025
Same author

Local Interactions Between Innate Immune Signaling, Microbiota, and Bile Acids Drive the Development of Duodenal Adenomas.

Cellular and molecular gastroenterology and hepatology·2025

Related Experiment Video

Updated: Dec 13, 2025

Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example
08:45

Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example

Published on: October 24, 2012

15.0K

Machine learning-based estimation of cognitive performance using regional brain MRI markers: the Northern Manhattan

Michelle R Caunca1,2,3, Lily Wang1, Ying Kuen Cheung4

  • 1Department of Public Health Sciences, Miller School of Medicine, University of Miami, Miami, FL, USA.

Brain Imaging and Behavior
|August 3, 2020
PubMed
Summary
This summary is machine-generated.

Simple models predict cognition well, but structural brain MRI markers are more useful for understanding causes than for prediction in older adults.

Keywords:
BiomarkersBrain agingCognitive agingMachine learning

More Related Videos

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

49.1K
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
14:27

Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data

Published on: June 26, 2013

16.0K

Related Experiment Videos

Last Updated: Dec 13, 2025

Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example
08:45

Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example

Published on: October 24, 2012

15.0K
Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

49.1K
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
14:27

Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data

Published on: June 26, 2013

16.0K

Area of Science:

  • Neuroscience
  • Cognitive Science
  • Medical Imaging

Background:

  • Machine learning and neuroimaging are increasingly used to predict cognitive function.
  • Comparisons between complex neuroimaging models and simpler models using basic variables are limited.
  • Regularization methods can identify key brain regions associated with cognition.

Purpose of the Study:

  • To compare the predictive performance of cognitive models using basic variables versus those including MRI markers.
  • To determine if structural brain MRI markers improve cognitive prediction compared to simpler models.
  • To explore the utility of MRI markers for etiological insights versus prediction.

Main Methods:

  • Used data from the Northern Manhattan Study cohort of older adults.
  • Compared three regression models: basic (sociodemographics, APOE ε4), basic + MRI, and MRI-only.
  • Employed machine learning techniques: elastic net, support vector regression, random forest, principal components regression.
  • Assessed model performance using RMSE, MAE, R², and 5-fold cross-validation.
  • Validated model significance against random biomarker datasets.

Main Results:

  • Basic models explained 31-38% of cognitive variance.
  • Adding MRI markers to basic models did not enhance cognitive estimation.
  • MRI-only models significantly outperformed random biomarker models (P < .05).
  • Identified relevant and novel regions-of-interest using MRI markers.

Conclusions:

  • Structural brain MRI markers may offer more etiological insights than predictive power for domain-specific cognition.
  • Simple models incorporating sociodemographics and APOE ε4 are effective for cognitive estimation.
  • Machine learning with MRI markers can identify potentially relevant brain regions, even if not improving prediction accuracy over basic models.