Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

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

Proton therapy range uncertainty reduction using vendor-agnostic tissue characterization on a virtual photon-counting CT head scan.

Medical physics·2026

Same author

Photon-Counting Detector vs. Energy-Integrating Detector CT Myelography for Detecting CSF-Venous Fistulas in Spontaneous Intracranial Hypotension: Substantially Lower Radiation Dose.

AJNR. American journal of neuroradiology·2026

Same author

Optimizing energy settings in CdTe, CZT, and Si photon-counting CT for material separation and detection.

Physics in medicine and biology·2026

Same author

Precise Lung Density Quantification with a Physics-based CT Harmonizer.

Radiology. Cardiothoracic imaging·2026

Same author

Utility of the virtual imaging trials methodology for objective characterization of AI systems and training data.

Journal of medical imaging (Bellingham, Wash.)·2026

Same author

Truth-based physics informed estimation of material composition in spectral CT in terms of density and effective atomic number.

Physics in medicine and biology·2026

Search research articles

Related Experiment Video

Updated: Jul 31, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Harmonizing CT Images via Physics-based Deep Neural Networks.

Mojtaba Zarei^1,2, Saman Sotoudeh-Paima^1,2, Cindy McCabe¹

¹Center for Virtual Imaging Trials, Carl E. Ravin Advanced Imaging Laboratories, Department of Radiology, Duke University School of Medicine.

Proceedings of Spie--The International Society for Optical Engineering

|May 3, 2023

Summary

This summary is machine-generated.

This study developed a physics-based deep neural network to harmonize computed tomography (CT) images, reducing variability in radiomics and biomarker quantification. The method improves the accuracy of medical image analysis for better disease assessment.

More Related Videos

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function

Published on: April 12, 2024

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

Related Experiment Videos

Last Updated: Jul 31, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function

Published on: April 12, 2024

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

Area of Science:

Medical Imaging
Artificial Intelligence
Radiomics

Background:

Image variations and biases in computed tomography (CT) hinder accurate quantification of imaging biomarkers.
Variability in CT reconstruction kernels and dose levels complicates radiomics and biomarker analysis.

Purpose of the Study:

To reduce variability in CT quantifications for radiomics and biomarkers.
To harmonize different CT scan renditions into a ground-truth-like image using physics-based deep neural networks (DNNs).

Main Methods:

Developed a generative adversarial network (GAN) where the generator incorporates the scanner's modulation transfer function (MTF).
Utilized a virtual imaging trial (VIT) platform with computational models (XCAT) and a CT simulator (DukeSim) for training.
Trained the model on CT images with varying dose levels (20 and 100 mAs) and reconstruction kernels.

Main Results:

Harmonized test set images achieved a structural similarity index of 0.95±0.1, normalized mean squared error of 10.2±1.5%, and peak signal-to-noise ratio of 31.8±1.5 dB.
Demonstrated more precise quantification of emphysema-based imaging biomarkers, including LAA-950, Perc15, and Lung mass.

Conclusions:

Physics-based DNNs, specifically GANs informed by MTF, can effectively harmonize CT images.
The proposed framework reduces variability and improves the precision of radiomics and biomarker quantifications, aiding in more accurate disease assessment.