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

Distinct profiles of circulating lymphocytes reflect clinical and serological variations in idiopathic inflammatory myopathies.

Reumatologia·2026
Same author

Beyond anti-SSA and anti-SSB. The clinical significance of new antibodies in the diagnosis of Sjögren disease.

Rheumatology international·2026
Same author

Updated Treatment of Non-Steroidal Anti-Inflammatory Drug-Exacerbated Respiratory Disease: How to Decide on Aspirin Therapy After Desensitization or Biologics? When? How? An EAACI Task Force Report.

Allergy·2026
Same author

Different faces of anti-melanoma differentiation-associated gene 5 antibody-positive myositis.

Reumatologia·2026
Same author

Prolonged myopathy and musculoskeletal symptoms following filgrastim in a 43-Year-Old female stem cell donor: a case-based review.

Rheumatology international·2025
Same author

Risk factors of nociplastic pain in patients with autoimmune arthritis: web-based cross-sectional survey of patients.

Rheumatology international·2025

Related Experiment Video

Updated: Jun 5, 2025

A Reproducible Computerized Method for Quantitation of Capillary Density using Nailfold Capillaroscopy
05:17

A Reproducible Computerized Method for Quantitation of Capillary Density using Nailfold Capillaroscopy

Published on: October 27, 2015

8.9K

Automatic assessment of nailfold capillaroscopy software: a pilot study.

Olga Elżbieta Brzezińska1, Krzysztof Andrzej Rychlicki-Kicior2, Joanna Samanta Makowska1

  • 1Department of Rheumatology, Medical University of Lodz, Poland.

Reumatologia
|December 16, 2024
PubMed
Summary

This study validates new software for automatic nailfold capillaroscopy image analysis. The tool accurately classifies images and counts capillaries, improving diagnosis for systemic sclerosis spectrum diseases.

Keywords:
artificial intelligencecapillaroscopydeep learningmicroangiopathy

More Related Videos

Non-Invasive Visualization of Nailbed Microvascular Morphology in Mice Using Capillaroscopy
05:06

Non-Invasive Visualization of Nailbed Microvascular Morphology in Mice Using Capillaroscopy

Published on: February 28, 2025

189
Non-invasive Assessment of Microvascular and Endothelial Function
05:41

Non-invasive Assessment of Microvascular and Endothelial Function

Published on: January 29, 2013

16.2K

Related Experiment Videos

Last Updated: Jun 5, 2025

A Reproducible Computerized Method for Quantitation of Capillary Density using Nailfold Capillaroscopy
05:17

A Reproducible Computerized Method for Quantitation of Capillary Density using Nailfold Capillaroscopy

Published on: October 27, 2015

8.9K
Non-Invasive Visualization of Nailbed Microvascular Morphology in Mice Using Capillaroscopy
05:06

Non-Invasive Visualization of Nailbed Microvascular Morphology in Mice Using Capillaroscopy

Published on: February 28, 2025

189
Non-invasive Assessment of Microvascular and Endothelial Function
05:41

Non-invasive Assessment of Microvascular and Endothelial Function

Published on: January 29, 2013

16.2K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence in Medicine
  • Rheumatology

Background:

  • Nailfold capillaroscopy is crucial for diagnosing systemic sclerosis spectrum diseases.
  • Manual image assessment is time-consuming and subjective, hindering comparative studies.
  • Objective and automated analysis is needed to standardize capillaroscopy assessments.

Purpose of the Study:

  • To validate pilot software for automatic capillaroscopy image classification (normal vs. pathological).
  • To assess the software's capability for automatic capillary counting.
  • To improve the objectivity and efficiency of capillaroscopy analysis.

Main Methods:

  • Trained a ResNet-34 deep neural network using the fast.ai library on 200 capillaroscopic images.
  • Employed 10-fold cross-validation with Darknet-YoloV3 on a GPU-optimized environment.
  • Developed an additional mechanism for calculating capillaries per millimeter.

Main Results:

  • Achieved 89.0% sensitivity and 89.4% specificity in classifying images as normal or pathological (training group).
  • Validation sensitivity was 89.0% and specificity 86.9%.
  • Demonstrated 96.48% precision in counting capillaries per millimeter.

Conclusions:

  • Pilot software offers a useful tool for rapid, automatic capillaroscopy assessment.
  • The software enables quick classification and capillary counting.
  • Future development aims for comprehensive imaging analysis independent of examiner experience.