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

RACE - Rapid Amplification of cDNA Ends02:35

RACE - Rapid Amplification of cDNA Ends

6.3K
Rapid Amplification of cDNA Ends, or RACE, is one of the most effective methods to obtain a full-length cDNA from an mRNA sequence between a known internal region to the unknown sequence at the 5’ or 3’ end. The unknown region is cloned in the cDNA by a gene-specific primer that binds the known end, and a hybrid primer that attaches a predefined anchor sequence to the unknown end of the cDNA. The sequence in between is amplified by PCR with an anchor primer and a gene-specific...
6.3K
GTPases and their Regulation02:14

GTPases and their Regulation

2.2K
2.2K
The Ratio of X Chromosome to Autosomes02:45

The Ratio of X Chromosome to Autosomes

8.5K
In most organisms, sex is determined by the ratio of X and Y chromosomes. However, in some organisms, such as Drosophila and C.elegans, sex is determined by the ratio of the number of X chromosomes to the number of sets of autosomes. The Y chromosome in Drosophila is active but does not determine sex. It contains genes responsible for the production of sperms in adult flies.  
Normal male Drosophila has a ratio of one X chromosome to two sets of autosomes. In contrast, normal female...
8.5K
Punnett Squares01:00

Punnett Squares

114.4K
Overview
114.4K
In- and Out-Groups01:31

In- and Out-Groups

39.0K
People all belong to a gender, race, age, and social economic group. These groups provide a powerful source of our identity and self-esteem (Tajfel & Turner, 1979) and serve as our in-groups. An in-group is a group that we identify with or see ourselves as belonging to.
39.0K
Poisson's Ratio01:23

Poisson's Ratio

402
Poisson's ratio is a material property that indicates their stress response. It explains the connection between the elongation or compression a material undergoes in the direction of an applied force and the contraction or expansion it experiences perpendicular to that force. When a slender bar is loaded axially, it stretches in the direction of the force and contracts laterally. Poisson's ratio is the negative ratio of this lateral contraction to the axial elongation. The negative sign...
402

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Comparative Analysis of Large Language Model Performance in Appropriate Diagnostic Imaging Modality Selection.

Journal of the American College of Radiology : JACR·2026
Same author

Observations of Wart Clearance Following COVID-19 Vaccination: Coincidence or Missed Immunologic Signals?

Vaccines·2025
Same author

Radiology in the Modern Emergency Department: Growth, Growth, and More Growth.

Radiology·2025
Same author

State Policies and Facility Practices of IV Hydration Spas in the US.

JAMA internal medicine·2025
Same author

Use of FDA Expedited Programs and Development Timelines for New Molecular Entity Drug and New Biologic Approvals by Sponsor Experience.

Clinical pharmacology and therapeutics·2025
Same author

Harnessing Youth Engagement at Faith Based Organizations to Combat Hypertension.

Preventing chronic disease·2025
Same journal

Comment on "Automated detection of superior mesenteric artery occlusion on post-contrast CT using a 3D deep learning model".

Clinical imaging·2026
Same journal

Visibility patterns of architectural distortion on synthesized 2D vs. digital breast tomosynthesis and their diagnostic implications.

Clinical imaging·2026
Same journal

Diagnostic performance of CT-based node-RADS for detecting metastatic lymph nodes in melanoma and comparison with short-axis and roundness index.

Clinical imaging·2026
Same journal

Identifying hepatocellular carcinoma patients at risk of early non-response after first-cycle transarterial chemoembolization: A reproducible machine learning study.

Clinical imaging·2026
Same journal

Comment on "The importance of FDG avidity in incidental thyroid nodules on FDG PET/CT".

Clinical imaging·2026
Same journal

Lipiodol in embolization therapy: Evolution from a diagnostic to a therapeutic agent.

Clinical imaging·2026
See all related articles

Related Experiment Video

Updated: Jun 29, 2025

Generalized Psychophysiological Interaction PPI Analysis of Memory Related Connectivity in Individuals at Genetic Risk for Alzheimer's Disease
09:38

Generalized Psychophysiological Interaction PPI Analysis of Memory Related Connectivity in Individuals at Genetic Risk for Alzheimer's Disease

Published on: November 14, 2017

14.9K

Even with ChatGPT, race matters.

Kanhai S Amin1, Howard P Forman2, Melissa A Davis2

  • 1Yale College, New Haven, CT, USA.

Clinical Imaging
|March 29, 2024
PubMed
Summary
This summary is machine-generated.

Large language models like ChatGPT show racial bias in simplifying medical reports. Readability varied significantly across racial groups, indicating a need for vigilance in AI applications.

Keywords:
ChatGPTHealth equityImplicit biasLarge language modelsRadiology report

More Related Videos

Using the Race Model Inequality to Quantify Behavioral Multisensory Integration Effects
08:13

Using the Race Model Inequality to Quantify Behavioral Multisensory Integration Effects

Published on: May 10, 2019

6.3K
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

554

Related Experiment Videos

Last Updated: Jun 29, 2025

Generalized Psychophysiological Interaction PPI Analysis of Memory Related Connectivity in Individuals at Genetic Risk for Alzheimer's Disease
09:38

Generalized Psychophysiological Interaction PPI Analysis of Memory Related Connectivity in Individuals at Genetic Risk for Alzheimer's Disease

Published on: November 14, 2017

14.9K
Using the Race Model Inequality to Quantify Behavioral Multisensory Integration Effects
08:13

Using the Race Model Inequality to Quantify Behavioral Multisensory Integration Effects

Published on: May 10, 2019

6.3K
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

554

Area of Science:

  • Artificial Intelligence in Medicine
  • Natural Language Processing

Background:

  • Large language models (LLMs) like ChatGPT are increasingly used in various applications.
  • It is critical to assess LLMs for potential perpetuation of racial inequities before widespread adoption.

Purpose of the Study:

  • To investigate whether ChatGPT-3.5 and ChatGPT-4 exhibit racial bias when simplifying radiology reports.
  • To compare the readability of simplified reports across different racial classifications.

Main Methods:

  • ChatGPT-3.5 and ChatGPT-4 were prompted to simplify 750 radiology reports.
  • The prompt included context of five major U.S. census racial classifications.
  • Readability scores of the simplified reports were calculated and compared across racial groups.

Main Results:

  • Statistically significant differences in readability were observed for both ChatGPT models based on racial context.
  • ChatGPT-3.5 outputs for White and Asian patients had higher reading grade levels compared to Black or African American and American Indian or Alaska Native patients.
  • ChatGPT-4 outputs for Asian patients had a higher reading grade level compared to American Indian or Alaska Native and Native Hawaiian or other Pacific Islander patients.

Conclusions:

  • The study found alarming differences in LLM outputs based on racial classifications, contrary to expectations.
  • These findings highlight the necessity for the medical community to remain vigilant against biased AI outputs.
  • Ensuring unbiased and non-harmful outputs from LLMs in healthcare is paramount.