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

Rapeseed root phospholipid metabolism orchestrates low phosphorus-induced microbiome changes and the interaction with beneficial Massilia.

Plant communications·2026
Same author

A Multimodal Measurement System for Quantifying Time-Dependent Suture Tension Relaxation and Wound Closure Strength.

Annals of biomedical engineering·2026
Same author

A General Kinematic Brain Injury Criterion Combining Translational and Rotational Effects.

Annals of biomedical engineering·2026
Same author

12-hydroxylauric acid, a novel growth regulator, promotes plant organ development.

Journal of integrative plant biology·2026
Same author

Targeting OXCT1 with the methyl donor S-adenosylmethionine as a therapeutic strategy for cerebral cavernous malformations.

International journal of biological macromolecules·2026
Same author

OsBOR1 Mediating Boron Efflux Controls Panicle Development via ROS Homeostasis in Rice.

Journal of agricultural and food chemistry·2026

Related Experiment Video

Updated: Jul 5, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

405

Anatomically Guided PET Image Reconstruction Using Conditional Weakly-Supervised Multi-Task Learning Integrating

Bao Yang, Kuang Gong, Huafeng Liu

    IEEE Transactions on Medical Imaging
    |January 19, 2024
    PubMed
    Summary
    This summary is machine-generated.

    This study introduces a new weakly-supervised multi-task learning strategy for positron emission tomography (PET) reconstruction. The method significantly reduces noise and improves image accuracy, enhancing diagnostic capabilities in PET imaging.

    More Related Videos

    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

    2.8K
    Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
    14:08

    Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

    Published on: April 13, 2013

    42.6K

    Related Experiment Videos

    Last Updated: Jul 5, 2025

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
    04:48

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

    Published on: July 5, 2024

    405
    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

    2.8K
    Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
    14:08

    Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

    Published on: April 13, 2013

    42.6K

    Area of Science:

    • Medical Imaging
    • Computer Vision
    • Machine Learning

    Background:

    • Positron Emission Tomography (PET) imaging faces challenges with high-quality training labels, limiting reconstruction accuracy.
    • Existing weakly-supervised methods for PET reconstruction can suffer from intrinsic variance and noise.
    • Improving the accuracy and generalizability of PET reconstruction models is crucial for clinical applications.

    Purpose of the Study:

    • To develop an improved weakly-supervised method for PET image reconstruction.
    • To suppress noise and enhance the accuracy and generalizability of PET reconstruction models.
    • To introduce an auxiliary anatomical task to regularize the main PET reconstruction task.

    Main Methods:

    • Proposed a conditional weakly-supervised multi-task learning (MTL) strategy for PET reconstruction.
    • Devised a novel multi-channel self-attention (MCSA) module to optimize feature sharing and capture dependencies.
    • Evaluated the method on NEMA phantom and clinical whole-body PET datasets.

    Main Results:

    • Achieved significant noise reduction (~50.0%) on phantom data compared to Maximum Likelihood (ML) reconstruction.
    • Demonstrated substantial noise reduction in patient studies (67.3% in liver, 35.5% in lung).
    • Showcased consistently small biases in tumor imaging and effective feature abstraction with MCSA.

    Conclusions:

    • The proposed MTL strategy with MCSA enhances PET reconstruction by reducing noise and improving accuracy.
    • The auxiliary task effectively integrates anatomical information, outperforming methods with only anatomical loss.
    • The developed approach offers superior performance in noise/contrast tradeoff and generalizability for PET imaging.