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

Chronic Treatment with Renin-Angiotensin System Inhibitors at Hospital Admission is Associated with Improved Reperfusion and Mortality Among STEMI Patients Undergoing Mechanical Reperfusion: Insight from the ISACS-STEMI COVID-19 Registry.

Current vascular pharmacology·2026
Same author

Development and Implementation of a Machine Learning Model for Prediction of Surgical Case Duration.

Joint Commission journal on quality and patient safety·2026
Same author

Comparison of [<sup>111</sup>In]octreotide Scintigraphy and [<sup>68</sup>Ga]DOTATOC PET in Gastroenteropancreatic Neuroendocrine Tumors: Concordance of Krenning Scores and Implications for Peptide Receptor Radionuclide Therapy Eligibility.

Neuroendocrinology·2026
Same author

Optimizing <sup>131</sup>I-mIBG Dosimetry: Validation of Simplified Time-Point and Segmentation Approaches.

Nuclear medicine and molecular imaging·2026
Same author

PHYTO-PET - Imaging plant physiology on a long-axial field-of-view PET scanner.

European journal of nuclear medicine and molecular imaging·2026
Same author

Reply: A Call for High-Level Evidence Before Routine Implementation of Posttreatment SPECT.

Journal of nuclear medicine : official publication, Society of Nuclear Medicine·2026

Related Experiment Video

Updated: Sep 26, 2025

Fat-Water Phantoms for Magnetic Resonance Imaging Validation: A Flexible and Scalable Protocol
07:59

Fat-Water Phantoms for Magnetic Resonance Imaging Validation: A Flexible and Scalable Protocol

Published on: September 7, 2018

11.6K

Noise-Based Image Harmonization Significantly Increases Repeatability and Reproducibility of Radiomics Features in

Harald Keller1,2, Tina Shek2, Brandon Driscoll2

  • 1Department of Radiation Oncology, University of Toronto, Toronto, ON M5T 1P5, Canada.

Tomography (Ann Arbor, Mich.)
|April 21, 2022
PubMed
Summary
This summary is machine-generated.

This study shows a simple image harmonization technique significantly improves the robustness and agreement of radiomics features derived from Positron Emission Tomography (PET) images, crucial for multicenter clinical studies.

Keywords:
PET radiomics featuresfeature agreementimage harmonizationrepeatabilityreproducibility

More Related Videos

Author Spotlight: Standardizing Mouse In Vivo PET Imaging with Body Conforming Molds and Automated Analysis
07:45

Author Spotlight: Standardizing Mouse In Vivo PET Imaging with Body Conforming Molds and Automated Analysis

Published on: October 25, 2024

504
Construction of a Preclinical Multimodality Phantom Using Tissue-mimicking Materials for Quality Assurance in Tumor Size Measurement
06:33

Construction of a Preclinical Multimodality Phantom Using Tissue-mimicking Materials for Quality Assurance in Tumor Size Measurement

Published on: July 29, 2013

11.4K

Related Experiment Videos

Last Updated: Sep 26, 2025

Fat-Water Phantoms for Magnetic Resonance Imaging Validation: A Flexible and Scalable Protocol
07:59

Fat-Water Phantoms for Magnetic Resonance Imaging Validation: A Flexible and Scalable Protocol

Published on: September 7, 2018

11.6K
Author Spotlight: Standardizing Mouse In Vivo PET Imaging with Body Conforming Molds and Automated Analysis
07:45

Author Spotlight: Standardizing Mouse In Vivo PET Imaging with Body Conforming Molds and Automated Analysis

Published on: October 25, 2024

504
Construction of a Preclinical Multimodality Phantom Using Tissue-mimicking Materials for Quality Assurance in Tumor Size Measurement
06:33

Construction of a Preclinical Multimodality Phantom Using Tissue-mimicking Materials for Quality Assurance in Tumor Size Measurement

Published on: July 29, 2013

11.4K

Area of Science:

  • Medical Imaging
  • Radiomics
  • Quantitative Imaging

Background:

  • Radiomics features derived from Positron Emission Tomography (PET) images are sensitive to image noise, impacting their robustness in multicenter clinical studies.
  • Standardizing image acquisition and reconstruction protocols across institutions is challenging, necessitating effective harmonization strategies.

Purpose of the Study:

  • To evaluate the efficacy of a simple contrast-to-noise ratio (CNR) equalization harmonization technique.
  • To assess the impact of this harmonization strategy on the robustness and agreement of radiomics features.
  • To identify suitable radiomics features for predictive modeling in multicenter PET studies.

Main Methods:

  • A texture pattern phantom was scanned on 10 PET scanners across 7 institutions using varied protocols.
  • An image harmonization technique based on equalizing contrast-to-noise ratio (CNR) was applied.
  • Reproducibility and repeatability studies were conducted, measuring feature agreement using intraclass correlation coefficient (ICC).

Main Results:

  • In repeatability studies, 81/93 features showed lower ICC with higher image noise.
  • The harmonized dataset significantly improved feature agreement for 5 out of 6 investigated feature classes compared to the standard dataset.
  • High feature agreement in the harmonized dataset correlated with higher sensitivity to texture patterns for three feature classes.

Conclusions:

  • A simple CNR-based image harmonization strategy effectively enhances radiomics feature robustness and agreement in multicenter PET studies.
  • This technique can mitigate the impact of image noise and inter-scanner variability.
  • The findings suggest a method for selecting robust and sensitive radiomics features for predictive model development.