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

Quantitative computerized analysis of diffuse lung disease in high-resolution computed tomography.

Yoshikazu Uchiyama1, Shigehiko Katsuragawa, Hiroyuki Abe

  • 1Kurt Rossmann Laboratories for Radiologic Image Research, Department of Radiology, The University of Chicago, 5841 S. Maryland Ave., Chicago, Illinois 60637, USA. yoshi@apex.bsd.uchicago.edu

Medical Physics
|October 8, 2003
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

Poly(ethylene glycol)-block-poly(glycidyl methacrylate) with oligoamine side chains as efficient gene vectors.

Macromolecular bioscience·2009
Same author

A thermochromic thin film based on host-guest interactions in a layered double hydroxide.

Langmuir : the ACS journal of surfaces and colloids·2009
Same author

Influence of dynamic load on friction behavior of human articular cartilage, stainless steel and polyvinyl alcohol hydrogel as artificial cartilage.

Journal of materials science. Materials in medicine·2009
Same author

[Effect of siRNA targeting c-Myc and VEGF on human colorectal cancer cells].

Zhonghua wei chang wai ke za zhi = Chinese journal of gastrointestinal surgery·2009
Same author

Tumor-specific, hypoxia-regulated, WW domain-containing oxidoreductase-expressing adenovirus inhibits human non-small cell lung cancer growth in vivo.

Human gene therapy·2009
Same author

Triosephosphate isomerase and peroxiredoxin 6, two novel serum markers for human lung squamous cell carcinoma.

Cancer science·2009

An automated system accurately detects diffuse lung diseases in high-resolution computed tomography (HRCT) images. This artificial intelligence tool aids radiologists in diagnosing various lung patterns with high sensitivity and specificity.

Area of Science:

  • Radiology
  • Medical Imaging
  • Artificial Intelligence

Background:

  • Diffuse lung diseases are challenging to diagnose using high-resolution computed tomography (HRCT).
  • Accurate characterization of lung patterns is crucial for effective patient management.

Purpose of the Study:

  • To develop and evaluate an automated computerized scheme for detecting and characterizing diffuse lung diseases on HRCT images.
  • To assess the performance of artificial neural networks (ANNs) in distinguishing between normal and abnormal lung patterns.

Main Methods:

  • A database of 315 HRCT images from 105 patients was curated, including six distinct diffuse lung disease patterns.
  • Radiologist-annotated "gold standard" data was used for training and validation.
  • Image segmentation, feature extraction (six physical measures), and ANNs were employed for pattern recognition.

Related Experiment Videos

Main Results:

  • The computerized method achieved high sensitivity for detecting specific patterns: 99.2% for ground-glass opacities, 100% for reticular/linear opacities, 88.0% for nodular opacities, 100% for honeycombing, 95.8% for emphysematous change, and 100% for consolidation.
  • Specificity for detecting normal regions of interest (ROIs) was 88.1%.

Conclusions:

  • The developed automated system demonstrates significant accuracy in identifying and characterizing diffuse lung diseases on HRCT.
  • This AI-powered tool shows potential as an aid to radiologists, improving diagnostic efficiency and accuracy for diffuse lung conditions.