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

Positron Emission Tomography01:29

Positron Emission Tomography

6.0K
Positron emission tomography (PET) is a medical imaging technique involving radiopharmaceuticals — substances that emit short-lived radiation. Although the first PET scanner was introduced in 1961, it took 15 more years before radiopharmaceuticals were combined with the technique and revolutionized its potential.
One of the main requirements of a PET scan is a positron-emitting radioisotope, which is produced in a cyclotron and then attached to a substance used by the part of the body...
6.0K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Image optimization for low-dose <sup>18</sup>F-FDG breast PET/MRI using deep learning: a pilot study.

EJNMMI physics·2026
Same author

Improved attenuation correction registration for FDG PET/CT images using data-driven gating (DDG)-based motion match.

EJNMMI physics·2026
Same author

Whole-body Dynamic [18F]FDG-PET/CT in Giant Cell Arteritis and Polymyalgia Rheumatica.

Clinical nuclear medicine·2026
Same author

Deep learning-enhanced data-driven gating improves FDG PET/CT clinical image quality.

EJNMMI physics·2026
Same author

Deep learning-enhanced digital-BGO versus TOF PET/CT: comparative assessment of detection, quantitation, and overall image quality.

EJNMMI physics·2025
Same author

Smartphone-based prediction of dopaminergic deficit in prodromal and manifest Parkinson's disease.

NPJ digital medicine·2025

Related Experiment Video

Updated: Sep 24, 2025

Multi-Tracer Studies of Brain Oxygen and Glucose Metabolism Using a Time-of-Flight Positron Emission Tomography-Computed Tomography Scanner
08:36

Multi-Tracer Studies of Brain Oxygen and Glucose Metabolism Using a Time-of-Flight Positron Emission Tomography-Computed Tomography Scanner

Published on: June 7, 2024

405

Deep learning-based time-of-flight (ToF) image enhancement of non-ToF PET scans.

Abolfazl Mehranian1, Scott D Wollenweber2, Matthew D Walker3

  • 1GE Healthcare, Big Data Institute, University of Oxford, Oxford, UK.

European Journal of Nuclear Medicine and Molecular Imaging
|May 4, 2022
PubMed
Summary

Deep learning models enhance non-time-of-flight (non-ToF) PET images to achieve time-of-flight (ToF) quality. This deep learning-based image enhancement (DL-ToF) improves quantitative accuracy and diagnostic confidence, offering a viable alternative to ToF reconstruction.

Keywords:
Deep neural networksImage qualityPETTime of flight

More Related Videos

Radiotracer Administration for High Temporal Resolution Positron Emission Tomography of the Human Brain: Application to FDG-fPET
09:03

Radiotracer Administration for High Temporal Resolution Positron Emission Tomography of the Human Brain: Application to FDG-fPET

Published on: October 22, 2019

10.3K
Creating Dynamic Images of Short-lived Dopamine Fluctuations with lp-ntPET: Dopamine Movies of Cigarette Smoking
14:21

Creating Dynamic Images of Short-lived Dopamine Fluctuations with lp-ntPET: Dopamine Movies of Cigarette Smoking

Published on: August 6, 2013

18.4K

Related Experiment Videos

Last Updated: Sep 24, 2025

Multi-Tracer Studies of Brain Oxygen and Glucose Metabolism Using a Time-of-Flight Positron Emission Tomography-Computed Tomography Scanner
08:36

Multi-Tracer Studies of Brain Oxygen and Glucose Metabolism Using a Time-of-Flight Positron Emission Tomography-Computed Tomography Scanner

Published on: June 7, 2024

405
Radiotracer Administration for High Temporal Resolution Positron Emission Tomography of the Human Brain: Application to FDG-fPET
09:03

Radiotracer Administration for High Temporal Resolution Positron Emission Tomography of the Human Brain: Application to FDG-fPET

Published on: October 22, 2019

10.3K
Creating Dynamic Images of Short-lived Dopamine Fluctuations with lp-ntPET: Dopamine Movies of Cigarette Smoking
14:21

Creating Dynamic Images of Short-lived Dopamine Fluctuations with lp-ntPET: Dopamine Movies of Cigarette Smoking

Published on: August 6, 2013

18.4K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence in Medicine
  • Radiology

Background:

  • Positron Emission Tomography (PET) imaging is crucial for disease diagnosis and monitoring.
  • Time-of-flight (ToF) reconstruction significantly improves PET image quality but requires specific hardware.
  • Non-ToF PET reconstruction can lead to quantitative inaccuracies and lower diagnostic confidence.

Purpose of the Study:

  • To develop and evaluate deep learning models for enhancing non-ToF PET images to achieve ToF-level image quality.
  • To improve the quantitative accuracy and diagnostic confidence of PET images reconstructed without ToF.
  • To assess the performance of deep learning-based ToF (DL-ToF) image enhancement.

Main Methods:

  • Utilized 273 [18F]-FDG PET scans from GE Discovery MI ToF scanners.
  • Reconstructed PET data using block-sequential-regularised-expectation-maximisation (BSREM) with and without ToF.
  • Trained three DL-ToF models (low, medium, high strength) to convert non-ToF to ToF images.
  • Objectively evaluated SUV accuracy in lesions and normal organs; subjectively assessed by radiologists for lesion detectability, confidence, and image quality.

Main Results:

  • DL-ToF models significantly reduced SUVmax differences in lesions compared to non-ToF, with the medium strength model showing a 1.7 ± 24% difference versus ToF-BSREM.
  • SUVmean differences in normal lung and liver regions were minimal across DL-ToF strengths, approaching ToF-BSREM values.
  • Radiologist scoring indicated DL-ToF medium and high models achieved diagnostic confidence comparable to or higher than standard ToF images, with DL-ToF medium scoring highest (4.1/5).
  • Visual assessment confirmed DL-ToF improved feature sharpness and convergence towards ToF reconstruction.

Conclusions:

  • Deep learning models can effectively generate ToF-equivalent PET image quality from non-ToF data.
  • DL-ToF enhancement offers a promising method to improve quantitative accuracy and diagnostic confidence without requiring ToF reconstruction.
  • The developed DL-ToF models are generalizable and can potentially be applied to non-ToF images from various PET/CT scanners, including BGO-based systems.