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 17, 2026

A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules
10:26

A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules

Published on: May 19, 2023

Pneumoconiosis screening and classification using deep learning models.

Meiqi Liu1, Zenas Huang2, Michal Borek2

  • 1Department of Statistics and Probability, Michigan State University, East Lansing, Michigan, USA.

Occupational and Environmental Medicine
|July 15, 2026
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

Development and validation of an E2 based indirect ELISA for serological detection of porcine Getah virus.

Veterinary research communications·2026
Same author

Thermal Ablation Therapy Integrated with Microenvironment Regulation and Anti-Inflammatory Effects for Choroidal Neovascularization.

ACS applied materials & interfaces·2026
Same author

Single-cell spatial profiling identifies a mucosal-like epithelial signature in Hidradenitis suppurativa tunnels.

JCI insight·2026
Same author

Quantitative Analysis of Cellular Ferroptosis Using Dual-Channel Colorimetry Based on Gold Nanoparticle Catalytic Activity.

Applied spectroscopy·2026
Same author

Wool-Derived Biochar via Composite Modification: Preparation and Enhanced Adsorption Performance for Methylene Blue.

Langmuir : the ACS journal of surfaces and colloids·2026
Same author

A green self-assembled hydrogel developed from glycyrrhizic acid and berberine for shelf-life extension and flavor quality preservation of strawberry fruits.

Food chemistry·2026

Deep learning models show promise for pneumoconiosis screening, performing comparably to certified B-readers on chest radiographs. Further research is needed to improve performance on diverse datasets and enhance generalizability.

Area of Science:

  • Radiology
  • Artificial Intelligence
  • Occupational Health

Background:

  • Pneumoconiosis poses a significant risk to workers in dusty industries.
  • Accurate screening and classification of pneumoconiosis are crucial for early intervention and management.
  • Current diagnostic methods rely on expert interpretation of chest radiographs, which can be subjective and time-consuming.

Purpose of the Study:

  • To develop and evaluate deep learning (DL) models for pneumoconiosis screening and classification.
  • To compare the performance of DL models against certified B-readers.
  • To assess the clinical utility of DL in diagnosing pneumoconiosis from chest radiographs.

Main Methods:

  • A retrospective study utilizing B-reader labeled pneumoconiosis lung images from US workers.
Keywords:
Artificial intelligenceCoal MiningLung Diseases, InterstitialOccupational Health

More Related Videos

Three-Dimensional Reconstruction for the Whole Lung with Early Multiple Pulmonary Nodules
07:53

Three-Dimensional Reconstruction for the Whole Lung with Early Multiple Pulmonary Nodules

Published on: October 13, 2023

Related Experiment Videos

Last Updated: Jul 17, 2026

A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules
10:26

A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules

Published on: May 19, 2023

Three-Dimensional Reconstruction for the Whole Lung with Early Multiple Pulmonary Nodules
07:53

Three-Dimensional Reconstruction for the Whole Lung with Early Multiple Pulmonary Nodules

Published on: October 13, 2023

  • Development and evaluation of DL models for four classification tasks: presence of abnormalities, parenchymal/pleural abnormalities, small opacity grading, and progressive massive fibrosis.
  • Independent testing using in-distribution and out-of-distribution datasets.

    • Main Results:

    • DL models achieved high accuracy (0.91-0.93) on in-distribution testing for tasks 1-3, comparable or superior to B-readers.
    • For progressive massive fibrosis classification (task 4), DL performance (accuracy 0.83) was similar to B-readers (accuracy 0.86).
    • Performance significantly decreased on the out-of-distribution test set, indicating challenges in generalization.

    Conclusions:

    • DL models demonstrate comparable performance to certified B-readers for pneumoconiosis screening and classification on in-distribution data.
    • DL holds potential to augment current pneumoconiosis diagnostic workflows.
    • Improving DL model generalizability across diverse datasets is essential for widespread clinical adoption.