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: Jul 23, 2025

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
12:18

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

7.6K

Detecting shortcut learning for fair medical AI using shortcut testing.

Alexander Brown1, Nenad Tomasev2, Jan Freyberg3

  • 1UCL Institute of Child Health, London, England.

Nature Communications
|July 18, 2023
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

Author Correction: The STARD-AI reporting guideline for diagnostic accuracy studies using artificial intelligence.

Nature medicine·2026
Same author

Accelerating scientific discovery with Co-Scientist.

ArXiv·2026
Same author

Towards Conversational AI for Disease Management.

Nature·2026
Same author

Surgical versus Non-Surgical Management of High-Grade Pediatric Pancreatic Trauma: A National TQIP Analysis Stratified by Hemodynamic Stability.

European journal of pediatric surgery : official journal of Austrian Association of Pediatric Surgery ... [et al] = Zeitschrift fur Kinderchirurgie·2026
Same author

Overutilization of helicopter emergency medical services compared to ground transport in moderate-severe penetrating abdominal trauma & associated outcomes.

Injury·2026
Same author

AI-Discovered Cognitive Models Reveal Novel Insights into Human and Animal Learning.

bioRxiv : the preprint server for biology·2026

Machine learning (ML) fairness is crucial in healthcare to prevent health disparities. This study introduces a method to detect shortcut learning, a cause of unfairness in clinical ML models, ensuring equitable AI applications.

Area of Science:

  • Artificial Intelligence in Medicine
  • Clinical Decision Support Systems
  • Health Equity Research

Background:

  • Machine learning (ML) offers significant potential for healthcare advancement.
  • Ensuring ML fairness is vital to prevent exacerbating health disparities.
  • Algorithmic unfairness can arise from shortcut learning, where models use spurious correlations.

Purpose of the Study:

  • To develop and validate a method for detecting shortcut learning in clinical ML.
  • To investigate the role of shortcut learning in ML model unfairness across different medical domains.
  • To differentiate shortcut learning from other causes of unfairness in medical AI.

Main Methods:

  • Utilized multitask learning to directly test for shortcut learning in ML models.

More Related Videos

One Dimensional Turing-Like Handshake Test for Motor Intelligence
14:05

One Dimensional Turing-Like Handshake Test for Motor Intelligence

Published on: December 15, 2010

26.9K
Assessment of Mouse Judgment Bias through an Olfactory Digging Task
12:10

Assessment of Mouse Judgment Bias through an Olfactory Digging Task

Published on: March 4, 2022

2.7K

Related Experiment Videos

Last Updated: Jul 23, 2025

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
12:18

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

7.6K
One Dimensional Turing-Like Handshake Test for Motor Intelligence
14:05

One Dimensional Turing-Like Handshake Test for Motor Intelligence

Published on: December 15, 2010

26.9K
Assessment of Mouse Judgment Bias through an Olfactory Digging Task
12:10

Assessment of Mouse Judgment Bias through an Olfactory Digging Task

Published on: March 4, 2022

2.7K
  • Applied the developed method to clinical tasks in radiology and dermatology.
  • Evaluated the performance differences of ML models across population subgroups.

    • Main Results:

    • Demonstrated a novel method for identifying shortcut learning in clinical ML applications.
    • Revealed instances where shortcut learning was and was not the cause of model unfairness.
    • Highlighted the complexity of achieving fairness in medical AI systems.

    Conclusions:

    • Shortcut learning is a key factor to consider in medical AI fairness.
    • A comprehensive approach is necessary for effective fairness mitigation in healthcare AI.
    • Distinguishing shortcut learning is essential for targeted interventions to ensure equitable AI deployment.