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

Deep Learning-Based Segmentation of Cryo-Electron Tomograms10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

10.6K
This is a method for training a multi-slice U-Net for multi-class segmentation of cryo-electron tomograms using a portion of one tomogram as a training input. We describe how to infer this network to other tomograms and how to extract segmentations for further analyses, such as subtomogram averaging and filament...
10.6K
Assessing Metabolic Neuromodulation Induced by Acute Deep Brain Stimulation in Rats Using In Vivo FDG-PET03:47

Assessing Metabolic Neuromodulation Induced by Acute Deep Brain Stimulation in Rats Using In Vivo FDG-PET

240
Source: Casquero-Veiga, M. et al. In vivo Positron Emission Tomography to Reveal Activity Patterns Induced by Deep Brain Stimulation in Rats. J. Vis. Exp. (2022).This video demonstrates the metabolic effects of deep brain stimulation (DBS) using FDG-PET imaging in a rat model. It outlines the steps involved in FDG administration, DBS application, and PET/CT imaging to map brain regions with increased or suppressed metabolic...
240
High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning09:31

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning

3.5K
A high-speed and open-top ultraviolet photoacoustic microscope that can provide histological images intraoperatively for surgical margin analysis is demonstrated, including the system configuration, optical alignment, sample preparation, and experimental...
3.5K
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography04:48

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

3.4K
An object segmentation protocol for orbital computed tomography (CT) images is introduced. The methods of labeling the ground truth of orbital structures by using super-resolution, extracting the volume of interest from CT images, and modeling multi-label segmentation using 2D sequential U-Net for orbital CT images are explained for supervised...
3.4K
Simultaneous PET/MRI Imaging During Mouse Cerebral Hypoxia-ischemia10:35

Simultaneous PET/MRI Imaging During Mouse Cerebral Hypoxia-ischemia

12.7K
The method presented here uses simultaneous positron emission tomography and magnetic resonance imaging. In the cerebral hypoxia-ischemia model, dynamic changes in diffusion and glucose metabolism occur during and after injury. The evolving and irreproducible damage in this model necessitates simultaneous acquisition if meaningful multi-modal imaging data are to be...
12.7K
Superior Auto-Identification of Trypanosome Parasites by Using a Hybrid Deep-Learning Model08:20

Superior Auto-Identification of Trypanosome Parasites by Using a Hybrid Deep-Learning Model

2.5K
Worldwide medical blood parasites were automatically screened using simple steps on a low-code AI platform. The prospective diagnosis of blood films was improved by using an object detection and classification method in a hybrid deep learning model. The collaboration of active monitoring and well-trained models helps to identify hotspots of trypanosome...
2.5K

You might also read

Related Articles

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

Sort by
Same author

Corrigendum to "Photodynamic therapy produces enhanced efficacy of antitumor immunotherapy by simultaneously inducing intratumoral release of sorafenib" [Biomaterials 2020, 240, 119845].

Biomaterials·2026
Same author

From PINNs to PIKANs: recent advances in physics-informed machine learning.

Machine learning for computational science and engineering·2026
Same author

Interfacial Redox Decoupling via Amphiphilic Carbon Dots for Highly Efficient Biphasic Photocatalytic H<sub>2</sub>O<sub>2</sub> Production and Selective Benzyl Alcohol Oxidation.

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

[<sup>18</sup>F] PSMA-3Q PET/CT radiomics at 45% SUVmax threshold: predicting post-surgical ISUP grade ≥ 4 and extraprostatic extension for noninvasive stratification in prostate cancer management.

Cancer imaging : the official publication of the International Cancer Imaging Society·2026
Same author

Hydrology-driven soil microbial responses: Inundation effects on soil microbial community succession and co-occurrence network stability in the lakeshore zone of Qinghai Lake.

Journal of environmental management·2026
Same author

Nutrient fluxes are underestimated in the Qinghai Lake Basin, the largest endorheic basin of the Asian water tower.

Water research·2026
Same journal

Subcortical dopamine D<sub>2</sub> receptor availability and glucose metabolism in autism: a dual-tracer PET/MR study.

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

PET-based assessment of coronary microvascular dysfunction and its relation to plaque burden and vascular remodeling: integration with microvascular resistance reserve.

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

Efficacy of peptide receptor radionuclide therapy in patients with oligodiscordant gastroenteropancreatic neuroendocrine tumours.

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

PET imaging of atherosclerotic plaques using the stabilin-2-targeted tracer <sup>18</sup>F-S2P in preclinical models.

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

Glucagon-like peptide-1 receptor PET in indolent and aggressive insulinoma: Diagnostic performance, quantitative differences, and clinical implications.

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

AI-based tumor and organ volumetry predicts early progression after Y-90 radioembolization in liver-dominant metastatic disease.

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

Related Experiment Video

Updated: Jan 20, 2026

Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

10.6K

PET image denoising using unsupervised deep learning.

Jianan Cui1,2, Kuang Gong1,3, Ning Guo1,3

  • 1Center for Advanced Medical Computing and Analysis, Massachusetts General Hospital, 55 Fruit St, White 427, Boston, MA, 02114, USA.

European Journal of Nuclear Medicine and Molecular Imaging
|August 31, 2019
PubMed
Summary
This summary is machine-generated.

This study introduces an unsupervised deep learning method for positron emission tomography (PET) image denoising, significantly improving image quality by utilizing prior patient information. The novel approach effectively reduces noise while preserving crucial image details, outperforming existing methods.

Keywords:
Anatomical priorDeep neural networkDenoisingPosition emission tomographyUnsupervised deep learning

More Related Videos

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning
09:31

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning

Published on: April 28, 2022

3.5K
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

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

Published on: November 30, 2022

3.4K

Related Experiment Videos

Last Updated: Jan 20, 2026

Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

10.6K
High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning
09:31

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning

Published on: April 28, 2022

3.5K
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

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

Published on: November 30, 2022

3.4K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Image Processing

Background:

  • Positron emission tomography (PET) image quality is often degraded by physical factors, limiting diagnostic accuracy.
  • Denoising PET images is crucial for enhancing visualization and quantitative analysis.
  • Existing methods struggle to balance noise reduction with the preservation of fine image details.

Purpose of the Study:

  • To develop and evaluate an unsupervised deep learning framework for PET image denoising.
  • To leverage prior patient information within the same scan for image restoration.
  • To achieve superior denoising performance without requiring paired training data.

Main Methods:

  • An unsupervised deep learning network was designed where a prior high-quality image served as input and the noisy PET image as the training label.
  • The network learned intrinsic image structures to restore the noisy PET image.
  • Performance was validated using simulation data (BrainWeb phantom) and clinical datasets (10 patients for 68Ga-PRGD2 PET/CT, 30 patients for 18F-FDG PET/MR).
  • Comparison was made against Gaussian, non-local mean (NLM), BM4D, and Deep Decoder methods using contrast-to-noise ratio (CNR) improvements.

Main Results:

  • The proposed method demonstrated superior performance in the bias-variance tradeoff during simulation studies.
  • Clinical evaluation showed significant CNR improvement: 53.35% ± 21.78% for PET/CT and 46.80% ± 25.23% for PET/MR.
  • These results significantly outperformed reference methods, with P-values < 0.002 for PET/CT and < 0.0001 for PET/MR.
  • Visual inspection confirmed effective noise smoothing and detail recovery in restored images across all datasets.

Conclusions:

  • The unsupervised deep learning framework offers effective PET image restoration.
  • The proposed method significantly outperforms traditional and deep learning-based denoising techniques.
  • This approach holds promise for improving the diagnostic utility of PET imaging.