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

Comparison of deep learning and particle smoother EM methods for estimation of Rb-82 myocardial perfusion PET kinetic parameters.

Medical physics·2026
Same author

Harnessing the potential of γδ T cells through engineering and combination treatment for cancer therapies.

Nature communications·2026
Same author

Maximum likelihood estimation yields accurate line-of-response assignment for positron + prompt gamma ray events in multiplexed PET (mPET).

Biomedical physics & engineering express·2026
Same author

Influence of scintillation light confinement on depth-of-interaction measurement performance in a single-ended readout PET detector.

Physics in medicine and biology·2026
Same author

Intracranial meningioma at foramen magnum presenting as bow hunter's syndrome via V4 segment compression: illustrative case.

Journal of neurosurgery. Case lessons·2026
Same author

Context-Aware Transformer GAN for Direct Generation of Attenuation and Scatter Corrected PET Data.

IEEE transactions on radiation and plasma medical sciences·2026
Same journal

<sup>18</sup>F-Flurpiridaz PET in myocardial perfusion imaging: comparing effectiveness against traditional tracers.

American journal of nuclear medicine and molecular imaging·2026
Same journal

Positron emission tomography imaging of T-cell activity in cardiovascular disease and its emerging role in imaging adaptive immunity.

American journal of nuclear medicine and molecular imaging·2026
Same journal

Calibration and prediction of results after failed injection in SPECT renal dynamic imaging.

American journal of nuclear medicine and molecular imaging·2026
Same journal

Development and evaluation of a novel deuterated fluorine-18 PET tracer for MAO-B imaging.

American journal of nuclear medicine and molecular imaging·2026
Same journal

<sup>18</sup>F-NaF and <sup>18</sup>F-FDG PET/CT imaging of spinal instrumentation: case report and review of the literature.

American journal of nuclear medicine and molecular imaging·2026
Same journal

Bladder activity of different PSMA PET radioligands and impact of furosemide.

American journal of nuclear medicine and molecular imaging·2026
See all related articles

Related Experiment Video

Updated: Mar 24, 2026

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities
07:13

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

Published on: October 27, 2023

1.8K

Using multimodal PET+MR data as conditional generative adversarial network inputs improves pseudo-CT and attenuation

Jonathan Fisher1,2, Emily Anaya1,2, Garry Chinn2

  • 1Department of Electrical Engineering, Stanford University Stanford, CA, USA.

American Journal of Nuclear Medicine and Molecular Imaging
|March 23, 2026
PubMed
Summary
This summary is machine-generated.

This study demonstrates that combining positron emission tomography (PET) and magnetic resonance (MR) imaging data using a multi-modality conditional generative adversarial network (cGAN) significantly improves brain PET attenuation correction accuracy. The developed method outperforms single-modality approaches and existing clinical techniques.

Keywords:
MR-based attenuation correction (MRAC)PETPET/MRattenuation correction (AC)deep learning (DL)emission-based attenuation correction (EBAC)generative adversarial network (GAN)

More Related Videos

Human Brown Adipose Tissue Depots Automatically Segmented by Positron Emission Tomography/Computed Tomography and Registered Magnetic Resonance Images
09:21

Human Brown Adipose Tissue Depots Automatically Segmented by Positron Emission Tomography/Computed Tomography and Registered Magnetic Resonance Images

Published on: February 18, 2015

12.7K
Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function
02:09

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

Published on: April 12, 2024

1.1K

Related Experiment Videos

Last Updated: Mar 24, 2026

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities
07:13

Author Spotlight: An Efficient and Robust Software for Automated Fusion of Multiple Preclinical Imaging Modalities

Published on: October 27, 2023

1.8K
Human Brown Adipose Tissue Depots Automatically Segmented by Positron Emission Tomography/Computed Tomography and Registered Magnetic Resonance Images
09:21

Human Brown Adipose Tissue Depots Automatically Segmented by Positron Emission Tomography/Computed Tomography and Registered Magnetic Resonance Images

Published on: February 18, 2015

12.7K
Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function
02:09

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

Published on: April 12, 2024

1.1K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence in Healthcare
  • Radiological Physics

Background:

  • Accurate positron emission tomography (PET) imaging requires correction for photon attenuation, typically achieved using X-ray computed tomography (CT) data.
  • Estimating attenuation correction (AC) factors from magnetic resonance (MR) data in integrated PET/MR scanners is challenging due to the lack of direct attenuation information in MR images.
  • Conditional generative adversarial networks (cGANs) have shown potential for both emission-based and MR-based AC.

Purpose of the Study:

  • To investigate the efficacy of combining PET and MR data using a multi-modality cGAN for improved brain PET attenuation correction accuracy.
  • To compare the performance of the multi-modality cGAN against single-modality cGANs and a clinical atlas-based method.

Main Methods:

  • Thirty-five patients underwent same-day whole-body PET/MR and PET/CT scans.
  • Four cGANs were trained to generate pseudo-CT images from non-attenuation-corrected and non-scatter-corrected (NASC) PET and MR data, with one network utilizing multi-modality input.
  • Performance was evaluated using structural similarity index (SSIM) and dice similarity coefficients for pseudo-CTs and SSIM and peak signal-to-noise ratio (PSNR) for AC PET images.

Main Results:

  • The multi-modality cGAN generated significantly superior pseudo-CT images compared to single-modality cGANs, achieving higher SSIM and dice similarity coefficients.
  • All cGAN-based methods outperformed the clinical atlas-based method for AC PET image reconstruction.
  • The multi-modal cGAN achieved the highest quality AC PET images, with average SSIM of 0.9987±0.0001 and PSNR of 50.0±0.4.

Conclusions:

  • Combining PET and MR data with a multi-modality cGAN offers a promising approach for enhancing brain PET attenuation correction.
  • This advanced method yields more accurate AC PET images than single-modality cGANs and current clinical atlas-based techniques.
  • The findings suggest a potential for improved diagnostic accuracy in PET/MR imaging through advanced AI-driven AC methods.