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

4.1K
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...
4.1K
Brain Imaging01:14

Brain Imaging

225
Brain imaging technologies provide critical insights into both the structure and function of the human brain, enabling medical professionals and researchers to diagnose, study, and treat neurological disorders or psychiatric disorders more effectively.
These technologies include computerized axial tomography (CAT or CT scans), positron-emission tomography (PET scans),  magnetic resonance imaging (MRI),  functional magnetic resonance imaging (fMRI), and Transcranial Magnetic...
225

You might also read

Related Articles

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

Sort by
Same author

Drug-Induced Micelle-to-Supermicelle Transition in Cetyltrimethylammonium Bromide (CTAB): Comparative Study with Acetylenic Gemini Surfactant via Experimental and DFT Approaches.

Langmuir : the ACS journal of surfaces and colloids·2026
Same author

Direct Spirocyclization of Tryptamines with Dearomatized Phenols Enables Catalyst-Free Access to C3-Spirocyclic Indolenines.

Organic letters·2026
Same author

Controlling the Delayed Fluorescence-Phosphorescence Balance via Torsional Dynamics and Intramolecular Charge Transfer in Carbazole-Based Aminoboranes.

The journal of physical chemistry. A·2026
Same author

Isolation of Mo(V)-dithiolene-radical: an elusive intermediate species relevant to catalytic cycle of molybdenum cofactor.

Chemical communications (Cambridge, England)·2026
Same author

Aging beyond diagnosis: the MRI brain age gap across disorders.

GeroScience·2026
Same author

Structural Reorganization Drives Exciton Relaxation Pathways in Layered 2D Ruddlesden-Popper (RP) Perovskite BA<sub>2</sub>PbI<sub>4</sub>.

Small (Weinheim an der Bergstrasse, Germany)·2026

Related Experiment Video

Updated: Jun 21, 2025

MRI and PET in Mouse Models of Myocardial Infarction
10:46

MRI and PET in Mouse Models of Myocardial Infarction

Published on: December 19, 2013

11.7K

Image synthesis of interictal SPECT from MRI and PET using machine learning.

Azin Shokraei Fard1, David C Reutens1,2,3, Stuart C Ramsay2

  • 1Centre for Advanced Imaging, University of Queensland, Brisbane, QLD, Australia.

Frontiers in Neurology
|July 11, 2024
PubMed
Summary

Generative adversarial networks (GANs) can estimate SPECT images from MRI and PET scans. This method shows promise for reducing radiation exposure and scan frequency in epilepsy patient evaluations.

Keywords:
cross modality image estimationdeep learninggenerative adversarial networksimage synthesisinterictal SPECTpre-surgical evaluations of epilepsy

More Related Videos

Simultaneous PET/MRI Imaging During Mouse Cerebral Hypoxia-ischemia
10:35

Simultaneous PET/MRI Imaging During Mouse Cerebral Hypoxia-ischemia

Published on: September 20, 2015

12.3K
PET Imaging of Neuroinflammation Using [11C]DPA-713 in a Mouse Model of Ischemic Stroke
12:01

PET Imaging of Neuroinflammation Using [11C]DPA-713 in a Mouse Model of Ischemic Stroke

Published on: June 14, 2018

12.8K

Related Experiment Videos

Last Updated: Jun 21, 2025

MRI and PET in Mouse Models of Myocardial Infarction
10:46

MRI and PET in Mouse Models of Myocardial Infarction

Published on: December 19, 2013

11.7K
Simultaneous PET/MRI Imaging During Mouse Cerebral Hypoxia-ischemia
10:35

Simultaneous PET/MRI Imaging During Mouse Cerebral Hypoxia-ischemia

Published on: September 20, 2015

12.3K
PET Imaging of Neuroinflammation Using [11C]DPA-713 in a Mouse Model of Ischemic Stroke
12:01

PET Imaging of Neuroinflammation Using [11C]DPA-713 in a Mouse Model of Ischemic Stroke

Published on: June 14, 2018

12.8K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Radiology

Background:

  • Cross-modality image estimation using generative adversarial networks (GANs) is an emerging technique.
  • The application of GANs for estimating SPECT (Single-Photon Emission Computed Tomography) images from other modalities like MRI (Magnetic Resonance Imaging) and PET (Positron Emission Tomography) has not been previously explored.
  • This study investigates the feasibility of SPECT image estimation from MRI and PET and assesses the role of cross-modality image registration in GAN training.

Purpose of the Study:

  • To evaluate the estimation of SPECT images from MRI and PET using GANs.
  • To determine the necessity of cross-modality image registration for effective GAN training in this context.
  • To assess the impact of different input configurations (single-channel vs. multi-channel) and loss function modifications on SPECT image synthesis quality.

Main Methods:

  • Interictal SPECT images were synthesized from PET and MRI data using the Pix2pix GAN framework, with data from 48 epilepsy patients.
  • Images were converted to 3D isotropic resolution and prepared in both native and template spaces for training and testing.
  • The study evaluated SPECT estimation using single-channel and multi-channel inputs, and assessed the effect of incorporating the structural similarity index metric into the GAN's loss function.

Main Results:

  • High-quality synthetic SPECT images were successfully generated from both MRI and PET data.
  • Utilizing images in native space yielded a 5.4% average improvement in structural similarity index (SSIM) compared to images registered to template space.
  • While PET provided the best results, MRI also produced SPECT images of comparable quality; adding SSIM to the loss function did not enhance image quality.

Conclusions:

  • Synthesizing SPECT images from MRI or PET using GANs is feasible and produces high-quality results.
  • This approach has the potential to significantly reduce the number of required scans for epilepsy patient evaluation.
  • The methodology offers a pathway to decrease patient radiation exposure by leveraging existing MRI or PET data for SPECT estimation.