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 Videos

Machine learning "red dot": open-source, cloud, deep convolutional neural networks in chest radiograph binary

E J Yates1, L C Yates1, H Harvey2

  • 1Foundation Doctor, West Midlands, England, UK.

Clinical Radiology
|June 15, 2018
PubMed
Summary
This summary is machine-generated.

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

Blood pressure management for caesarean delivery under spinal anaesthesia: a UK multi-centre audit (2023).

International journal of obstetric anesthesia·2025
Same author

AI implementation in the UK landscape: Knowledge of AI governance, perceived challenges and opportunities, and ways forward for radiographers.

Radiography (London, England : 1995)·2024
Same author

Thoracic aortic aneurysm repair using the elephant trunk technique and associated complications.

Clinical radiology·2022
Same author

Outcomes and interventions in patients transported to hospital with ongoing CPR after out-of-hospital cardiac arrest - An observational study.

Resuscitation plus·2021
Same author

Introduction of an abbreviated breast MRI service in the UK as part of the BRAID trial: practicalities, challenges, and future directions.

Clinical radiology·2021
Same author

Reply to: Letter to the Editor re out-of-hospital cardiac arrest termination of resuscitation with ongoing CPR: An observational study.

Resuscitation·2018
Same journal

Where can we improve? An evaluation of discrepancies between radiology resident and consultant reports for noncontrast head computed tomography (NCHCT) scans in the emergency department at a tertiary care centre.

Clinical radiology·2026
Same journal

Proton beam therapy: neuroradiology considerations and challenges.

Clinical radiology·2026
Same journal

Weight-based contrast media volume reduction in lower limb computed tomography angiography: a prospective feasibility study.

Clinical radiology·2026
Same journal

Diagnostic accuracy of flat panel computed tomography for acute stroke: a systematic review and meta-analysis of diagnostic test accuracy studies.

Clinical radiology·2026
Same journal

Nuclear medicine techniques in gynaecological malignancy: a review of best practice.

Clinical radiology·2026
Same journal

Deep learning reconstruction dual-energy computed tomography for gastrointestinal system tumors: low-kiloelectron volt imaging vs routine imaging.

Clinical radiology·2026
See all related articles

A machine learning model was developed to classify chest radiographs, achieving 94.6% accuracy in identifying abnormalities. This AI tool can help prioritize radiologist workload for improved efficiency.

Area of Science:

  • Artificial Intelligence in Medical Imaging
  • Machine Learning for Diagnostic Support

Background:

  • Chest radiography interpretation is crucial for diagnosing various conditions.
  • Clinician workload can be optimized through efficient reporting prioritization.

Purpose of the Study:

  • To develop a machine learning (ML) model for binary classification of chest radiography abnormalities.
  • To create a retrospective tool for guiding clinician reporting prioritization.

Main Methods:

  • Utilized Tensorflow and the Inception deep convolutional neural network (CNN).
  • Retrained the final layer for binary normality classification on public datasets.
  • Trained on 47,644 images and validated on 5,505 unseen radiographs.

Main Results:

Related Experiment Videos

  • Achieved a final model accuracy of 94.6% on the unseen testing subset.
  • Reported sensitivity of 94.6% and specificity of 93.4%.
  • Obtained a positive predictive value (PPV) of 99.8% and an area under the curve (AUC) of 0.98.

Conclusions:

  • Demonstrated the successful application of an ML-based approach for chest radiograph classification.
  • The model shows potential for optimizing clinician workload in real-world settings.