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

A vendor-neutral functional MRI acquisition protocol for multi-site studies.

Aperture neuro·2026
Same author

Reply: A Call for High-Level Evidence Before Routine Implementation of Posttreatment SPECT.

Journal of nuclear medicine : official publication, Society of Nuclear Medicine·2026
Same author

Noise-aware adaptive diffusion sampling for accelerated knee MRI reconstruction.

Physics in medicine and biology·2026
Same author

Phantom- and simulation-based validation of combined diffusion relaxometry in ex vivo ADRD white matter.

bioRxiv : the preprint server for biology·2026
Same author

Smooth optimization using global and local low-rank regularizers.

SIAM journal on imaging sciences·2026
Same author

Bilevel Optimized Implicit Neural Representation for Scan-Specific Accelerated MRI Reconstruction.

IEEE transactions on medical imaging·2026

Related Experiment Video

Updated: Dec 13, 2025

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations
12:27

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

Published on: February 15, 2017

7.2K

Improved Low-Count Quantitative PET Reconstruction With an Iterative Neural Network.

Hongki Lim, Il Yong Chun, Yuni K Dewaraja

    IEEE Transactions on Medical Imaging
    |August 4, 2020
    PubMed
    Summary
    This summary is machine-generated.

    This study introduces a new deep learning method, BCD-Net, to improve low-count Positron Emission Tomography (PET) image reconstruction. BCD-Net enhances image quality by reducing noise and improving contrast, crucial for accurate diagnosis in treatments like Y-90 radioembolization.

    More Related Videos

    Author Spotlight: Optimizing Dendritic Spine Analysis for Balanced Manual and Automated Assessment in the Hippocampus CA1 Apical Dendrites
    07:45

    Author Spotlight: Optimizing Dendritic Spine Analysis for Balanced Manual and Automated Assessment in the Hippocampus CA1 Apical Dendrites

    Published on: September 27, 2024

    2.8K
    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

    4.1K

    Related Experiment Videos

    Last Updated: Dec 13, 2025

    Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations
    12:27

    Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

    Published on: February 15, 2017

    7.2K
    Author Spotlight: Optimizing Dendritic Spine Analysis for Balanced Manual and Automated Assessment in the Hippocampus CA1 Apical Dendrites
    07:45

    Author Spotlight: Optimizing Dendritic Spine Analysis for Balanced Manual and Automated Assessment in the Hippocampus CA1 Apical Dendrites

    Published on: September 27, 2024

    2.8K
    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

    4.1K

    Area of Science:

    • Medical Imaging
    • Nuclear Medicine
    • Artificial Intelligence

    Background:

    • Low-count Positron Emission Tomography (PET) imaging faces challenges due to high random event fractions from natural radioactivity, reducing signal-to-noise ratio (SNR).
    • Model-based image reconstruction (MBIR) requires effective regularization to control noise, as unregularized methods can amplify it with increased iterations.
    • Learned convolutional operators are emerging as advanced regularization techniques in MBIR.

    Purpose of the Study:

    • To modify and evaluate a deep learning-based iterative neural network, BCD-Net, for improved image reconstruction in low-count PET.
    • To assess the efficacy of BCD-Net in handling high random fractions and low true coincidence count-rates, particularly relevant for Y-90 PET imaging.
    • To compare the performance of BCD-Net against traditional regularization methods like Total Variation (TV) and Non-Local Means (NLM).

    Main Methods:

    • Modification of the BCD-Net iterative neural network architecture for PET MBIR.
    • Training and testing the BCD-Net using simulated XCAT phantom data mimicking low count-rates and high random fractions.
    • Validation using clinically relevant phantom measurement data with diverse activity distributions and count levels.
    • Comparison of BCD-Net with MBIR using TV and NLM regularizers.

    Main Results:

    • The trained BCD-Net significantly improved Contrast-to-Noise Ratio (CNR) and reduced Root Mean Square Error (RMSE) in reconstructed images compared to TV and NLM methods.
    • BCD-Net demonstrated successful generalization to unseen test data with different characteristics from the training data.
    • Performance improvements were validated on clinical phantom data, even with highly dissimilar training and testing datasets.

    Conclusions:

    • The proposed BCD-Net offers a superior regularization approach for low-count PET MBIR, effectively addressing challenges posed by high random fractions.
    • This deep learning-based method enhances image quality metrics (CNR, RMSE) crucial for accurate patient imaging and diagnosis.
    • BCD-Net shows robust generalization capabilities, making it a promising tool for clinical applications in Y-90 PET and similar low-count scenarios.