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

Split-Spectrum Amplitude-Decorrelation Optoretinography Detects Impaired Photoreceptor Function in Age-Related Macular Degeneration.

Ophthalmology science·2026
Same author

Adverse Events Associated With Midface Distractors: A Review of the Manufacturer and User Facility Device Experience (MAUDE) Database.

The Journal of craniofacial surgery·2026
Same author

In Vitro Antiviral Effects of Green-Lipped Mussel Oil and Low-Molecular-Weight Fucoidan on HSV, RSV, and SARS-CoV-2 Pseudovirus.

Biomedicines·2026
Same author

Drusen volume and reticular pseudodrusen volume from optical coherence tomography with deep learning as risk factors for progression to late age-related macular degeneration in eyes with reticular pseudodrusen and contralateral macular neovascularisation.

The British journal of ophthalmology·2026
Same author

Optical coherence tomography angiography: Principles and applications.

Handbook of clinical neurology·2026
Same author

Moderating Effects of Social Determinants of Health on Quality of Life in Expressive Writing Interventions Among Chinese American Breast Cancer Survivors.

Psycho-oncology·2026

Related Experiment Video

Updated: Jan 17, 2026

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

16.1K

Clinically Explainable Disease Diagnosis Based on Biomarker Activation Map.

Pengxiao Zang, Carol Wang, Tristan T Hormel

    IEEE Transactions on Bio-Medical Engineering
    |September 25, 2025
    PubMed
    Summary

    A new biomarker activation map (BAM) framework explains artificial intelligence (AI) disease diagnoses. This AI explainability tool highlights key biomarkers, improving clinical trust and adoption of AI diagnostic classifiers.

    More Related Videos

    Author Spotlight: Engineering Molecular Tools for Disease Detection and Imaging
    04:33

    Author Spotlight: Engineering Molecular Tools for Disease Detection and Imaging

    Published on: December 8, 2023

    1.4K
    Detection of Human Leukocyte Antigen Biomarkers in Breast Cancer Utilizing Label-free Biosensor Technology
    08:27

    Detection of Human Leukocyte Antigen Biomarkers in Breast Cancer Utilizing Label-free Biosensor Technology

    Published on: March 24, 2015

    15.3K

    Related Experiment Videos

    Last Updated: Jan 17, 2026

    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

    16.1K
    Author Spotlight: Engineering Molecular Tools for Disease Detection and Imaging
    04:33

    Author Spotlight: Engineering Molecular Tools for Disease Detection and Imaging

    Published on: December 8, 2023

    1.4K
    Detection of Human Leukocyte Antigen Biomarkers in Breast Cancer Utilizing Label-free Biosensor Technology
    08:27

    Detection of Human Leukocyte Antigen Biomarkers in Breast Cancer Utilizing Label-free Biosensor Technology

    Published on: March 24, 2015

    15.3K

    Area of Science:

    • Biomedical Informatics
    • Medical Imaging Analysis
    • Artificial Intelligence in Healthcare

    Background:

    • Artificial intelligence (AI) disease classifiers demonstrate expert-level diagnostic performance.
    • The
    • black box
    • nature of AI hinders real-world clinical adoption.
    • Explainability is crucial for integrating AI into clinical workflows.

    Purpose of the Study:

    • To introduce a novel biomarker activation map (BAM) generation framework.
    • To provide clinically meaningful explainability for AI-based disease classifiers.
    • To enhance the adoption and trustworthiness of AI diagnostic tools.

    Main Methods:

    • Developed a framework based on residual counterfactual explanation.
    • Generated counterfactual outputs to reverse classifier decisions.
    • Created BAMs as difference maps between counterfactuals and original inputs.
    • Validated BAMs across four diverse classifiers: AMD, diabetic retinopathy, brain tumors, and breast cancer.

    Main Results:

    • Highlighted regions in BAMs strongly correlated with manually identified disease biomarkers.
    • Demonstrated the framework's effectiveness across multiple imaging modalities (fundus photography, OCT angiography, MRI, CT).
    • BAMs provided intuitive visual explanations for AI diagnostic predictions.

    Conclusions:

    • The BAM framework significantly enhances the clinical applicability of AI disease classifiers.
    • Provides clinicians with understandable outputs to verify AI diagnostic decisions.
    • Facilitates greater trust and integration of AI tools in medical diagnostics.