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

Multi-modal segment anything model (mmSAM) for tumor segmentation in multi-tracer oncologic PET/CT.

EJNMMI physics·2026
Same author

A kernel space-based multidimensional sparse model for dynamic PET image denoising.

Physics in medicine and biology·2026
Same author

Evaluation of regularization factor and attenuation correction method of Q.Clear reconstruction for brain PET/MR in epileptic patients.

EJNMMI physics·2026
Same author

Revisiting the segmentation threshold for Lu-177 SPECT.

Medical physics·2026
Same author

One-stop evaluation using [<sup>68</sup>Ga]Ga-Pentixafor PET integrated with contrast-enhanced CT for visualization and localization of adrenal nodules in patients with primary aldosteronism.

Japanese journal of radiology·2026
Same author

Acquisition of K<sup>trans</sup> perfusion parameter maps from DCE-MRI in breast cancer using a deep learning approach.

Quantitative imaging in medicine and surgery·2026

Related Experiment Video

Updated: Jul 22, 2025

Retrospective Cardiac Gating with A Prototype Small-Animal X-ray Computed Tomograph
05:32

Retrospective Cardiac Gating with A Prototype Small-Animal X-ray Computed Tomograph

Published on: February 21, 2025

336

Transfer learning-based attenuation correction for static and dynamic cardiac PET using a generative adversarial

Hao Sun1,2,3,4, Fanghu Wang5, Yuling Yang1,3,4

  • 1School of Biomedical Engineering, Southern Medical University, 1023 Shatai Road, Guangzhou, 510515, China.

European Journal of Nuclear Medicine and Molecular Imaging
|July 20, 2023
PubMed
Summary

This study demonstrates a deep learning approach to generate accurate attenuation-corrected PET images from non-corrected images for cardiac [13N]ammonia PET scans. This method shows comparable results to traditional CT-based correction, improving image quality and myocardial blood flow assessment.

Keywords:
Attenuation correctionDeep learningMyocardial blood flowMyocardial perfusion PETTransfer learning

More Related Videos

Management of Respiratory Motion Artefacts in 18F-fluorodeoxyglucose Positron Emission Tomography using an Amplitude-Based Optimal Respiratory Gating Algorithm
06:53

Management of Respiratory Motion Artefacts in 18F-fluorodeoxyglucose Positron Emission Tomography using an Amplitude-Based Optimal Respiratory Gating Algorithm

Published on: July 23, 2020

5.6K
High-Resolution Cardiac Positron Emission Tomography/Computed Tomography for Small Animals
11:09

High-Resolution Cardiac Positron Emission Tomography/Computed Tomography for Small Animals

Published on: December 16, 2022

3.8K

Related Experiment Videos

Last Updated: Jul 22, 2025

Retrospective Cardiac Gating with A Prototype Small-Animal X-ray Computed Tomograph
05:32

Retrospective Cardiac Gating with A Prototype Small-Animal X-ray Computed Tomograph

Published on: February 21, 2025

336
Management of Respiratory Motion Artefacts in 18F-fluorodeoxyglucose Positron Emission Tomography using an Amplitude-Based Optimal Respiratory Gating Algorithm
06:53

Management of Respiratory Motion Artefacts in 18F-fluorodeoxyglucose Positron Emission Tomography using an Amplitude-Based Optimal Respiratory Gating Algorithm

Published on: July 23, 2020

5.6K
High-Resolution Cardiac Positron Emission Tomography/Computed Tomography for Small Animals
11:09

High-Resolution Cardiac Positron Emission Tomography/Computed Tomography for Small Animals

Published on: December 16, 2022

3.8K

Area of Science:

  • Medical Imaging
  • Nuclear Medicine
  • Artificial Intelligence

Background:

  • Accurate attenuation correction (AC) is crucial for quantitative analysis in Positron Emission Tomography (PET).
  • Traditional CT-based AC (CTAC) requires accurate CT data and can be affected by patient motion.
  • Developing automated AC methods is essential for improving efficiency and accuracy in cardiac PET imaging.

Purpose of the Study:

  • To demonstrate the feasibility of directly generating attenuation-corrected PET images from non-attenuation-corrected (NAC) PET images.
  • To apply a generative adversarial network (GAN) for creating attenuation-corrected images for [13N]ammonia myocardial perfusion (MP) PET.
  • To evaluate the performance of deep learning-based AC (DLAC) for both static and dynamic cardiac PET scans.

Main Methods:

  • A 3D pix2pix deep learning AC (DLAC) framework was developed using a U-net + ResNet generator and a CNN discriminator.
  • Paired static and dynamic NAC and CTAC PET images from 60 rest-only subjects were used for training static (S-DLAC) and dynamic (D-DLAC) models.
  • Transfer learning was employed to fine-tune the S-DLAC model into an improved dynamic model (D-DLAC-FT) for dynamic PET images.

Main Results:

  • The DLAC methods (S-DLAC, D-DLAC, D-DLAC-FT) demonstrated consistency with clinical CTAC in image quality and quantitative metrics.
  • S-DLAC showed a significantly higher correlation with static CTAC (R²=0.947) compared to static NAC (R²=0.654).
  • D-DLAC-FT achieved comparable myocardial blood flow (MBF) accuracy to D-DLAC, with low errors in both rest and stress-state subjects.

Conclusions:

  • The proposed DLAC methods achieve performance comparable to clinical CTAC for [13N]ammonia MP PET.
  • Deep learning-based attenuation correction shows significant promise for improving dynamic myocardial perfusion PET.
  • Transfer learning is a valuable technique for enhancing the performance of dynamic PET image analysis.