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

Radiological Investigation I: X-ray and CT01:30

Radiological Investigation I: X-ray and CT

Radiological investigations, including X-rays and computed tomography (CT) scans, are critical for diagnosing and evaluating various medical conditions. These imaging techniques provide valuable insights into the body's internal structures, aiding in the detection of abnormalities, assessment of disease progression, and development of treatment strategies. This article delves into two primary radiological investigations, chest X-rays and CT scans, outlining their purpose, procedures, and the...
Imaging Studies for Cardiovascular System III: X-Ray01:20

Imaging Studies for Cardiovascular System III: X-Ray

The most common cardiovascular diagnostic test is an X-ray. It produces images of the heart, blood vessels, and adjacent structures.
Definition and Purpose
An X-ray, or radiograph, is a non-invasive method that uses ionizing radiation to take images of internal structures. It is mainly used in cardiac imaging to examine the heart, lungs, and major blood vessels, aiming to identify abnormalities in the heart's size, shape, and position, such as heart failure, congenital defects, and vascular...
X-ray Imaging01:24

X-ray Imaging

German physicist Wilhelm Röntgen (1845–1923) was experimenting with electrical current when he discovered that a mysterious and invisible "ray" would pass through his flesh but leave an outline of his bones on a screen coated with a metal compound. In 1895, Röntgen made the first durable record of the internal parts of a living human: an "X-ray" image (as it came to be called) of his wife’s hand. Scientists worldwide quickly began their own experiments with X-rays, and by 1900, X-ray was widely...

You might also read

Related Articles

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

Sort by
Same author

A Pilot Study Protocol for AI-Assisted Interpretation of Chest X-rays for Pulmonary Abnormalities in Uganda.

Cureus·2026
Same author

Deception in clinical large language models: an under-recognised safety risk.

The Lancet. Digital health·2026
Same author

BRIDGE: benchmarking large language models for understanding real-world clinical practice texts.

Nature biomedical engineering·2026
Same author

An evidence gap map of digital health interventions for enhancing patient engagement in healthcare.

NPJ digital medicine·2026
Same author

Discriminating HFrEF vs HFpEF from chest radiographs: Mitigating demographic performance gaps via augmentation and multimodal fusion.

PLOS digital health·2026
Same author

Dialogues of delivery: a multilingual question-answer dataset for maternal healthcare in East African languages.

BMC research notes·2026

Related Experiment Videos

Impact of Exposure Parameters on Deep Learning Models in Chest Radiography and Implications for Deployment.

Han-Jay Shu1, Shun-Ting Chang1, Rodrigo Rosa Gameiro2,3

  • 1Department of Computer Science, National Tsing Hua University, Hsinchu, Taiwan.

Radiology. Artificial Intelligence
|June 24, 2026
PubMed
Summary

Deep learning models trained on chest X-rays may use radiographic exposure parameters as shortcuts, leading to biased results. Auditing exposure parameters can identify high-risk conditions before clinical use.

Related Experiment Videos

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Radiology

Background:

  • Deep learning models are increasingly used for analyzing chest radiographs (CXRs).
  • Concerns exist that these models may rely on unintended features, such as radiographic exposure parameters, leading to biased performance.
  • Understanding and mitigating these biases is crucial for reliable clinical deployment.

Purpose of the Study:

  • To determine if deep learning models trained on CXRs utilize radiographic exposure parameters as shortcut features.
  • To quantify the impact of these shortcut features on model performance under different exposure conditions.
  • To assess the utility of exposure-regimen screening for identifying potential biases.

Main Methods:

  • Retrospective analysis of 727,604 CXRs from three large datasets (MIMIC-CXR, MIDRC, EmoryCXR).
  • Extraction of exposure parameters (ExposureTime, XRayTubeCurrent, ExposureInuAs) from DICOM metadata.
  • Training and evaluation of models on CXRs for pneumothorax detection, COVID-19 diagnosis, and race classification under controlled and natural exposure regimes.

Main Results:

  • Significant performance declines were observed when models were evaluated on mismatched exposure distributions across all tasks (pneumothorax detection: ΔAUC = -0.38; COVID-19: ΔAUC = -0.33; race classification: ΔAUC = -0.09).
  • A priori screening identified high-risk exposure regimes within natural data distributions that correlated with reduced model performance.
  • Deep learning models demonstrated a reliance on exposure parameters, acting as shortcut features.

Conclusions:

  • Deep learning models trained on CXRs may exploit radiographic exposure parameters, introducing biases.
  • Exposure-regimen audits are recommended to identify and flag high-risk conditions prior to clinical deployment of AI models.
  • Ensuring model robustness against variations in imaging parameters is essential for trustworthy AI in radiology.