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

Reply to Wang et al.: A Clearer Clinical Role for ORACLE-IPF in IPF Prognostication.

American journal of respiratory and critical care medicine·2026
Same author

Association of biological age acceleration with muscle aging phenotype in young and middle-aged adults: a prospective cohort study.

GeroScience·2026
Same author

Enhanced biofuel productivity via interfacial synergy in robust core-shell silicotungstic acid@zeolitic imidazolate framework-8 nanocatalysts for canola oil (trans)esterification.

Bioresource technology·2026
Same author

Rheumatoid arthritis and risk of myocardial infarction: A nationwide cohort study.

European journal of preventive cardiology·2026
Same author

Physical Unclonable Function Based on 3D-NAND Flash Array Structure With Multi-Chip Implementation.

Small (Weinheim an der Bergstrasse, Germany)·2026
Same author

Comparative evaluation of generative AI models for chest radiograph report generation in the emergency department.

European radiology·2026

Related Experiment Video

Updated: Dec 31, 2025

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
05:49

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization

Published on: February 23, 2024

1.3K

Test-retest reproducibility of a deep learning-based automatic detection algorithm for the chest radiograph.

Hyungjin Kim1,2, Chang Min Park3,4,5, Jin Mo Goo1,2,6

  • 1Department of Radiology, Seoul National University College of Medicine, 101, Daehak-ro, Jongno-gu, Seoul, 03080, South Korea.

European Radiology
|January 5, 2020
PubMed
Summary

The deep learning-based automatic detection algorithm (DLAD) shows good reproducibility on chest radiographs. However, less conspicuous pulmonary nodules may lead to output fluctuations and potential misclassifications, impacting diagnostic accuracy.

Keywords:
Artificial intelligenceComputer-assisted radiographic image interpretationRadiographyReproducibility of resultsSolitary pulmonary nodule

More Related Videos

Three-Dimensional Phase Resolved Functional Lung Magnetic Resonance Imaging
10:44

Three-Dimensional Phase Resolved Functional Lung Magnetic Resonance Imaging

Published on: June 21, 2024

993
Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
08:05

Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia

Published on: December 19, 2020

14.6K

Related Experiment Videos

Last Updated: Dec 31, 2025

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
05:49

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization

Published on: February 23, 2024

1.3K
Three-Dimensional Phase Resolved Functional Lung Magnetic Resonance Imaging
10:44

Three-Dimensional Phase Resolved Functional Lung Magnetic Resonance Imaging

Published on: June 21, 2024

993
Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
08:05

Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia

Published on: December 19, 2020

14.6K

Area of Science:

  • Radiology
  • Artificial Intelligence
  • Medical Imaging

Background:

  • Deep learning-based automatic detection algorithms (DLAD) are increasingly used in medical imaging.
  • Assessing the reproducibility of these algorithms is crucial for clinical adoption.
  • Chest radiographs (CRs) are common imaging modalities for detecting pulmonary abnormalities.

Purpose of the Study:

  • To evaluate the test-retest reproducibility of a DLAD for pulmonary nodule detection using stationary CRs.
  • To identify factors influencing test-retest variations in DLAD performance.
  • To assess the robustness of DLAD to simulated image post-processing and positional changes.

Main Methods:

  • Retrospective analysis of preoperative CRs from 169 patients with resected pulmonary nodules.
  • Test-retest reproducibility assessed using median abnormality score differences, intraclass correlation coefficients (ICC), and 95% limits of agreement (LoA).
  • Univariable and multivariable analyses investigated factors associated with variation; simulated post-processing and positional changes were evaluated for robustness.

Main Results:

  • DLAD demonstrated good test-retest reproducibility with median abnormality score differences of 1-2% and ICCs ranging from 0.83 to 0.90.
  • Test-retest variation was negatively associated with solid portion size and nodule conspicuity.
  • DLAD was robust to simulated positional changes (ICC: 0.984–0.996) but less so to post-processing (ICC: 0.872–0.968).
  • A high-specificity cutoff (46%) led to discordant classifications in 8.9% of cases.

Conclusions:

  • The DLAD exhibits general robustness to test-retest variations in chest radiographs.
  • Pulmonary nodule conspicuity and solid portion size are key factors influencing DLAD output stability.
  • Careful consideration of classification cutoffs is necessary to mitigate misclassifications, especially for inconspicuous nodules.