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

Two-dimensional extrapolation methods for texture analysis on CT scans.

William F Sensakovic1, Adam Starkey, Samuel G Armato

  • 1Department of Radiology, The University of Chicago, 5841 South Maryland Avenue, Chicago, Illinois 60637, USA. wfsensak@uchicago.edu

Medical Physics
|October 12, 2007
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

Development and validation of artificial intelligence-assisted volumetric response criteria in pleural mesothelioma (ARTIMES): a retrospective, multicohort, multicentre study.

The Lancet. Oncology·2026
Same author

A pilot study of magnetic resonance fingerprinting and radiomics analysis in autosomal dominant polycystic kidney disease.

Kidney international·2026
Same author

Correction to: 2025 ACC/AHA/ASE/ASNC/SCCT/SCMR Advanced Training Statement on Advanced Cardiovascular Imaging: A Report of the ACC Competency Management Committee.

Circulation. Cardiovascular imaging·2026
Same author

Self-supervised out-of-distribution detection-Metal implants and other anomaly.

Medical physics·2026
Same author

2025 ACC/AHA/ASE/ASNC/SCCT/SCMR Advanced Training Statement on Advanced Cardiovascular Imaging: A Report of the ACC Competency Management Committee.

Journal of cardiovascular magnetic resonance : official journal of the Society for Cardiovascular Magnetic Resonance·2025
Same author

2025 ACC/AHA/ASE/ASNC/SCCT/SCMR Advanced Training Statement on Advanced Cardiovascular Imaging: A Report of the ACC Competency Management Committee.

Journal of the American College of Cardiology·2025
Same journal

Correction to "On the shape of the radiation survival curve in tumor spheroids: The role of oxygen heterogeneity".

Medical physics·2026
Same journal

Multi-view constrained semi-supervised vertebra detection for 3D ultrasound spine volume.

Medical physics·2026
Same journal

Accuracy of quantitative <sup>177</sup>Lu SPECT/CT imaging: A systematic review.

Medical physics·2026
Same journal

Physics-constrained dual-domain network for CBCT reconstruction from orthogonal X-rays in gynecologic radiotherapy.

Medical physics·2026
Same journal

Decomposition-based harmonization for quantitative PET imaging across scanners and radiotracers.

Medical physics·2026
Same journal

Development and evaluation of an in vivo dose-based monitoring system for electron FLASH radiation therapy.

Medical physics·2026
See all related articles

Texture analysis on medical images can be inaccurate with incomplete regions of interest (ROIs). This study found CLEAN deconvolution best corrects deficient ROIs, improving texture descriptor accuracy and visual similarity.

Area of Science:

  • Medical imaging analysis
  • Radiomics
  • Computational pathology

Background:

  • Texture analysis in medical imaging relies on complete regions of interest (ROIs).
  • Deficient ROIs, where tissue does not fully occupy the ROI, can compromise texture descriptor accuracy and computational speed.
  • Lung parenchyma analysis using thoracic CT scans often encounters such deficient ROIs.

Purpose of the Study:

  • To evaluate the impact of deficient ROIs on texture descriptors.
  • To assess the effectiveness of three extrapolation methods (mean fill, tiled fill, CLEAN deconvolution) in correcting deficient ROIs for texture analysis.
  • To determine the best method for improving accuracy and visual fidelity of texture descriptors.

Main Methods:

  • Applied 198 texture descriptors from five classes (statistical, Fourier, fractal, Laws' filtered) to lung parenchyma ROIs from ten patients' thoracic CT scans.

Related Experiment Videos

  • Calculated descriptors on original, deficient, and corrected ROIs using mean fill, tiled fill, and CLEAN deconvolution methods.
  • Statistically compared descriptor values across original, deficient, and corrected ROIs to assess accuracy and visual similarity.

    • Main Results:

    • Statistically significant differences were observed in 138 of 198 texture descriptors when calculated on deficient ROIs compared to original ROIs.
    • All three extrapolation methods showed statistically significant improvements in texture descriptor accuracy for certain descriptors.
    • CLEAN deconvolution demonstrated the most significant improvements across the greatest number of descriptors and yielded the best overall accuracy and visual similarity.

    Conclusions:

    • Deficient ROIs significantly impact texture descriptor accuracy in medical image analysis.
    • Extrapolation methods can correct for deficient ROIs, with CLEAN deconvolution proving most effective.
    • CLEAN deconvolution offers a robust solution for improving the reliability and visual quality of texture analysis in medical imaging, particularly for lung parenchyma CT scans.