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

Processing, Characterization and Applications of Ceramic Matrix Composites.

Materials (Basel, Switzerland)·2026
Same author

Anatomy-aware lymphoma lesion detection in whole-body PET/CT.

Frontiers in oncology·2026
Same author

Elasticity-guided tumor resection: applying biomechanical information to neurosurgical practice.

Acta neurochirurgica·2026
Same author

Chainsaw culture: Cladophialophora bantiana cerebral abscess after chainsaw activity in an immunocompetent host. Illustrative case.

Journal of neurosurgery. Case lessons·2026
Same author

Differential cholinergic white matter microstructural alterations in multiple system atrophy and Parkinson's disease.

Neurobiology of disease·2026
Same author

Impact of tractogram filtering and graph creation for structural connectomics in subjects with Parkinson's disease.

Frontiers in human neuroscience·2026
Same journal

Investigating the Neural Origins of Ear-EEG: A Correlation Study Using Scalp EEG Source Reconstruction.

NeuroImage·2026
Same journal

Hysteresis effects in visual and auditory perception and the comparison of underlying neural mechanisms - an EEG study.

NeuroImage·2026
Same journal

Short-term audio-tactile training affects cortical auditory speech-envelope tracking for incongruent but not congruent stimuli.

NeuroImage·2026
Same journal

Dissociable Neurocognitive Mechanisms of State and Trait Anxiety in Working Memory: Threat-Induced Alterations in Decision Dynamics and Attenuation of Large-Scale Network Reconfiguration.

NeuroImage·2026
Same journal

Neuro-Ocular Amyloid Characterization in Alzheimer's Disease via Cross-Site PET-MRI and Hierarchical Cross-Attention Driven Multimodal Representation Learning.

NeuroImage·2026
Same journal

Whole-brain network dynamics underlying intolerance of uncertainty.

NeuroImage·2026
See all related articles

Related Experiment Video

Updated: Jun 20, 2026

Application of Granger Causality Analysis of the Directed Functional Connection in Alzheimer's Disease and Mild Cognitive Impairment
08:43

Application of Granger Causality Analysis of the Directed Functional Connection in Alzheimer's Disease and Mild Cognitive Impairment

Published on: August 7, 2017

Explainable AI for MRI Alzheimer's disease classification: A comparative analysis.

Sanna Persson1, Hugo Svenberg1, Alexis Moscoso Rial2

  • 1Department of Biomedical Engineering and Health Systems, KTH Royal Institute of Technology, Stockholm, Sweden.

Neuroimage
|June 18, 2026
PubMed
Summary
This summary is machine-generated.

We developed a quantitative pipeline to explain Alzheimer's disease (AD) prediction models using explainable AI (XAI) methods. Different XAI approaches highlight distinct brain regions, offering complementary insights into AD pathology.

Keywords:
Alzheimer’s diseaseExplainable AIImage classificationMagnetic resonance imagingMultimodal neuroimaging

More Related Videos

Basics of Multivariate Analysis in Neuroimaging Data
06:35

Basics of Multivariate Analysis in Neuroimaging Data

Published on: July 24, 2010

DeepOmicsAE: Representing Signaling Modules in Alzheimer's Disease with Deep Learning Analysis of Proteomics, Metabolomics, and Clinical Data
09:47

DeepOmicsAE: Representing Signaling Modules in Alzheimer's Disease with Deep Learning Analysis of Proteomics, Metabolomics, and Clinical Data

Published on: December 15, 2023

Related Experiment Videos

Last Updated: Jun 20, 2026

Application of Granger Causality Analysis of the Directed Functional Connection in Alzheimer's Disease and Mild Cognitive Impairment
08:43

Application of Granger Causality Analysis of the Directed Functional Connection in Alzheimer's Disease and Mild Cognitive Impairment

Published on: August 7, 2017

Basics of Multivariate Analysis in Neuroimaging Data
06:35

Basics of Multivariate Analysis in Neuroimaging Data

Published on: July 24, 2010

DeepOmicsAE: Representing Signaling Modules in Alzheimer's Disease with Deep Learning Analysis of Proteomics, Metabolomics, and Clinical Data
09:47

DeepOmicsAE: Representing Signaling Modules in Alzheimer's Disease with Deep Learning Analysis of Proteomics, Metabolomics, and Clinical Data

Published on: December 15, 2023

Area of Science:

  • Neuroimaging
  • Artificial Intelligence
  • Computational Neuroscience

Background:

  • Alzheimer's disease (AD) prediction models require transparent explanations for clinical trust.
  • Post hoc explainable AI (XAI) methods offer insights into model decision-making but require quantitative validation.
  • Region-level analysis of neuroimaging data is crucial for understanding AD biomarkers.

Purpose of the Study:

  • To introduce and validate a quantitative pipeline for region-level explanations of an Alzheimer's prediction model.
  • To compare the performance of four different XAI methods (DeepLIFT, Layer Gradient×Activation, Occlusion sensitivity analysis, XGrad-CAM) in identifying AD-related brain regions.
  • To assess the fidelity and cross-modal correspondence of XAI-derived explanations with PET imaging data.

Main Methods:

  • Developed a quantitative pipeline mapping XAI explanations to Freesurfer anatomical regions.
  • Summarized explanations as the percentage of regional volume deemed relevant by each XAI method.
  • Validated the pipeline using cross-sectional and longitudinal ADNI cohorts, including cognitively normal (CN), AD, and mild cognitive impairment (MCI) subjects.
  • Evaluated XAI fidelity using perturbation analyses and correlated regional MRI relevance with amyloid and tau PET SUVRs.

Main Results:

  • XAI methods (except XGrad-CAM) identified hippocampus/amygdala as discriminative for CN, while AD-relevant regions were method-dependent.
  • Longitudinal analysis revealed increased caudate/putamen/pallidum relevance across most XAI methods for subjects converting to AD.
  • Attribution profiles in amyloid-positive MCI subjects were intermediate between AD and CN groups.
  • Weak correlations between XAI maps and PET data suggest MRI and PET provide complementary information.

Conclusions:

  • XAI results for AD classification are dependent on the specific method and disease class.
  • Different XAI methods provide complementary information, aiding end-users in understanding AD's biological underpinnings.
  • Applying multiple XAI methods is recommended for comprehensive AD classification and interpretation.