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 Video

Updated: Mar 14, 2026

Decoding Natural Behavior from Neuroethological Embedding
08:00

Decoding Natural Behavior from Neuroethological Embedding

Published on: October 3, 2025

833

Deep learning interpretability in neuroimaging: A comprehensive survey and methodological recommendations.

Md Mahfuzur Rahman1,2, Vince Calhoun2, Sergey Plis1,2

  • 1Department of Computer Science, Georgia State University, Atlanta, GA, United States.

Imaging Neuroscience (Cambridge, Mass.)
|March 13, 2026
PubMed
Summary

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

Group Joint ICA (gjICA): A Method for Multimodal Fusion of Concurrent EEG and fMRI Data.

Human brain mapping·2026
Same author

Aberrant recovery of timescale-aligned amplitude balance links to symptoms and cognition in schizophrenia.

Translational psychiatry·2026
Same author

Multimodal subspace independent vector analysis effectively captures latent relationships between brain structure and function.

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

Large-scale brain dynamics are organized by a directional coordination hierarchy.

bioRxiv : the preprint server for biology·2026
Same author

Structure-function coupling of large-scale cortical networks across the lifespan is spectrally specific.

Communications biology·2026
Same author

Modeling Complex Effects and Individual Variability in Multi-Paradigm fMRI with Nonlinear Mixed Models.

bioRxiv : the preprint server for biology·2026

Deep learning (DL) models offer powerful biomarker discovery in neuroimaging but face challenges due to their opacity. Improving DL interpretability is crucial for safe clinical integration in mental health.

Area of Science:

  • Neuroimaging
  • Artificial Intelligence
  • Mental Health

Background:

  • Deep learning (DL) models are increasingly used in neuroimaging for biomarker discovery.
  • DL models learn directly from raw data, bypassing traditional feature extraction.
  • The opacity of DL models hinders their clinical adoption in healthcare.

Purpose of the Study:

  • To review the literature on DL interpretability in neuroimaging.
  • To highlight challenges and strategies for DL deployment in mental health.
  • To foster trust and transparency in AI for clinical applications.

Main Methods:

  • Comprehensive literature review of DL interpretability.
  • Focus on applications within neuroimaging studies.
  • Analysis of strategies for enhancing model transparency.
Keywords:
brain dynamicsdeep learninginterpretabilityneuroimagingpsychiatric disorders

More Related Videos

Whole-Brain Single-Cell Imaging and Analysis of Intact Neonatal Mouse Brains Using MRI, Tissue Clearing, and Light-Sheet Microscopy
08:49

Whole-Brain Single-Cell Imaging and Analysis of Intact Neonatal Mouse Brains Using MRI, Tissue Clearing, and Light-Sheet Microscopy

Published on: August 1, 2022

4.3K

Related Experiment Videos

Last Updated: Mar 14, 2026

Decoding Natural Behavior from Neuroethological Embedding
08:00

Decoding Natural Behavior from Neuroethological Embedding

Published on: October 3, 2025

833
Whole-Brain Single-Cell Imaging and Analysis of Intact Neonatal Mouse Brains Using MRI, Tissue Clearing, and Light-Sheet Microscopy
08:49

Whole-Brain Single-Cell Imaging and Analysis of Intact Neonatal Mouse Brains Using MRI, Tissue Clearing, and Light-Sheet Microscopy

Published on: August 1, 2022

4.3K

Main Results:

  • DL shows promise for analyzing complex neuroimaging data.
  • Model interpretability remains a significant barrier to clinical translation.
  • Transparency is essential for trust in safety-critical healthcare domains.

Conclusions:

  • Interpretable DL is vital for integrating AI into clinical practice for mental disorders.
  • Further research is needed to develop and validate DL interpretability methods.
  • Bridging the gap between DL capabilities and clinical needs requires a focus on transparency.