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

Towards Automated Quality Assurance: Integrating Deep Learning and Classical ML into the Digital Radiography

Hsuan-Yu Chen1,2, Cheng-Fu Chou3, Sheng-Hung Liao4

  • 1Department of Orthopedic Surgery, College of Medicine, National Taiwan University, No.7, Chung-Shan South Road, Zhong-Zheng District, Taipei 100, Taiwan.

Diagnostics (Basel, Switzerland)
|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

Controllability and trajectory controls of an underactuated flapping-wing aircraft steered by center of gravity.

Bioinspiration & biomimetics·2026
Same author

Targeting glutamine metabolism combined with a nanoradioenhancer in radioresistant hepatocellular carcinoma.

Nanoscale·2026
Same author

Delta-like 1 homolog (DLK1) regulates cancer stemness and chemoresistance of ovarian cancer by CD44/CD133 upregulation.

The Journal of pathology·2026
Same author

A New Way to Engineer Cell Sheets for Articular Cartilage Regeneration.

Journal of functional biomaterials·2025
Same author

medspacyV: a graphical user interface for the open source medspaCy natural language processing package.

JAMIA open·2025
Same author

Correction: Chang et al. Fabrication of Stromal Cell-Derived Factor-1 Contained in Gelatin/Hyaluronate Copolymer Mixed with Hydroxyapatite for Use in Traumatic Bone Defects. <i>Micromachines</i> 2021, <i>12</i>, 822.

Micromachines·2025

A deep learning system automates quality control for Lumbar Spinal Digital Radiographs (LSDR), improving image evaluation and diagnostic reliability. This AI approach enhances efficiency and accuracy in medical imaging analysis.

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Radiography

Background:

  • Manual review of Lumbar Spinal Digital Radiographs (LSDR) is time-consuming and prone to variability.
  • Automating quality control for LSDR is crucial for improving diagnostic accuracy and workflow efficiency.

Purpose of the Study:

  • To develop and evaluate a deep learning-based automated quality control system for LSDR.
  • To enhance the evaluation of LSDR and reduce reliance on manual reviews.

Main Methods:

  • A retrospective study using a deep learning workflow involving image segmentation, feature extraction, and classification.
  • Four U-Net-based models were assessed, with Attention U-Net with weight map selected for superior performance.
  • An XGBoost classifier utilized extracted features (brightness, contrast, anatomical positioning) for image classification.
Keywords:
automateddeep learninglumbar spine X-raypatient safetyquality control

Related Experiment Videos

Main Results:

  • Attention U-Net with weighted attention achieved high mean intersection over union (mIoU) scores for AP and LAT views.
  • The XGBoost classifier demonstrated excellent performance (AUC ~0.9) in classifying LSDR as qualified or unqualified.
  • The deep learning approach effectively handled class imbalances, outperforming traditional machine learning models.

Conclusions:

  • The developed automated quality control system shows significant potential for improving LSDR image quality.
  • This AI-driven system can enhance diagnostic reliability and optimize clinical workflow efficiency.
  • Automated quality control systems are vital for advancing modern medical imaging practices.