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: Oct 19, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

695

Going Beyond Saliency Maps: Training Deep Models to Interpret Deep Models.

Zixuan Liu1, Ehsan Adeli2,3, Kilian M Pohl2,4

  • 1Department of Electrical Engineering, Stanford University, Stanford, CA 94305, USA.

Information Processing in Medical Imaging : Proceedings of the ... Conference
|September 22, 2021
PubMed
Summary
This summary is machine-generated.

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

Leveraging Machine Learning to Advance Alcohol Research: Current Applications, Challenges, and Opportunities.

Alcohol research : current reviews·2026
Same author

A framework of digital biomarkers for neurodegenerative diseases.

Nature reviews bioengineering·2026
Same author

SocialGen: Modeling Multi-Human Social Interaction with Language Models.

Proceedings. International Conference on 3D Vision·2026
Same author

Mapping Individualized Developmental Imbalance in Youth and Its Association with Psychopathology.

bioRxiv : the preprint server for biology·2026
Same author

Using deep learning to identify brain networks mediating cognitive and motor impairments in alcohol use disorder.

Translational psychiatry·2026
Same author

A generalized synthetic control algorithm for sparse functional data.

bioRxiv : the preprint server for biology·2026
Same journal

Enhancing Alzheimer's Diagnosis: Leveraging Anatomical Landmarks in Graph Convolutional Neural Networks on Tetrahedral Meshes.

Information processing in medical imaging : proceedings of the ... conference·2026
Same journal

Cycle-Consistent Zero-Shot Through-Plane Super-Resolution for Anisotropic Head MRI.

Information processing in medical imaging : proceedings of the ... conference·2026
Same journal

Brightness-Invariant Tracking Estimation in Tagged MRI.

Information processing in medical imaging : proceedings of the ... conference·2025
Same journal

Multi-View and Multi-Scale Alignment for Contrastive Language-Image Pre-training in Mammography.

Information processing in medical imaging : proceedings of the ... conference·2025
Same journal

Using Multiple Instance Learning to Build Multimodal Representations.

Information processing in medical imaging : proceedings of the ... conference·2025
Same journal

mSPD-NN: A Geometrically Aware Neural Framework for Biomarker Discovery from Functional Connectomics Manifolds.

Information processing in medical imaging : proceedings of the ... conference·2024
See all related articles

This study introduces a new method for interpreting deep learning models in neuroimaging. It visualizes disease patterns in MRIs, offering clearer insights into brain disorders than traditional saliency maps.

Area of Science:

  • Neuroimaging
  • Artificial Intelligence
  • Computational Neuroscience

Background:

  • Deep learning models are crucial for analyzing neuroimaging data but often lack interpretability.
  • Existing methods like saliency maps provide limited, non-intuitive insights into disease-related morphological changes.

Purpose of the Study:

  • To develop a more interpretable method for deep learning classifiers in neuroimaging.
  • To visualize specific disease-related patterns in MRI scans for better understanding of brain disorders.

Main Methods:

  • Proposed a novel image-to-image translation approach using simulator networks trained with conditional convolution.
  • Developed a method to inject or remove disease patterns into MRI scans via warping operations.
  • Utilized Jacobian determinants of the warping field for visualization and interpretation.

More Related Videos

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis
05:41

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

9.6K

Related Experiment Videos

Last Updated: Oct 19, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

695
A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis
05:41

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

9.6K

Main Results:

  • The proposed method successfully visualized meaningful patterns associated with Alzheimer's disease and alcohol dependence.
  • Simulations revealed disease-specific morphological changes, offering greater interpretability than traditional saliency maps.
  • The approach demonstrated robustness across synthetic and real neuroimaging datasets.

Conclusions:

  • The simulator network approach enhances the interpretability of deep learning models in neuroimaging.
  • This method provides human-understandable visualizations of disease effects, advancing the study of brain disorders.
  • The technique offers a promising avenue for understanding complex neurological conditions through AI.