Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

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

[<sup>177</sup>Lu]Lu-edotreotide versus everolimus for gastroenteropancreatic neuroendocrine tumours (COMPETE): a phase 3, multicentre, randomised, open-label, superiority trial.

Lancet (London, England)·2026

Same author

Is there already value in "Total Body PET" imaging in Neurology?

The British journal of radiology·2026

Same author

Generative Consistency Models for Estimation of Kinetic Parametric Image Posteriors in Total-Body PET.

IEEE transactions on medical imaging·2026

Same author

Diffusion models for medical image reconstruction.

BJR artificial intelligence·2026

Same author

Large Language Models for Patient Education: Insights from Cross-Lingual Expert Evaluation in Radioligand Therapy for Neuroendocrine Tumors.

Journal of cancer education : the official journal of the American Association for Cancer Education·2026

Same author

The Dementias Platform UK PET/MR harmonisation and test-retest study: assessment of PET repeatability and reproducibility across the national network.

European journal of nuclear medicine and molecular imaging·2026

Same journal

Impact of DOI Capability on Detector Performance: A Comparative Study of DOI and Non-DOI Detectors for High-Resolution and Sensitivity Organ-Specific PET Inserts.

IEEE transactions on radiation and plasma medical sciences·2026

Same journal

Evaluation of a PET Insert for Trimodal Imaging: A Step Toward PET/MRI-Guided Focused Ultrasound.

IEEE transactions on radiation and plasma medical sciences·2026

Same journal

CHEMONO: a Cherenkov-Only Monolithic Detector for PGI in Proton Range Verification.

IEEE transactions on radiation and plasma medical sciences·2026

Same journal

A Virtual Clinical Trial to Detect Changes in Tumor Uptake with PET using Lesion Embedding.

IEEE transactions on radiation and plasma medical sciences·2026

Same journal

Quantitative evaluation of spatially-variant deformations recovered by deep learning on clinical-like breast lesions.

IEEE transactions on radiation and plasma medical sciences·2026

Same journal

Impact of detector parameters and image resolution modeling on dedicated brain PET imaging.

IEEE transactions on radiation and plasma medical sciences·2026

See all related articles

Search research articles

Related Experiment Video

Updated: Jun 10, 2025

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Deep Convolutional Backbone Comparison for Automated PET Image Quality Assessment.

Jessica B Hopson¹, Anthime Flaus², Colm J McGinnity²

¹Department of Biomedical Engineering, King's College London.

IEEE Transactions on Radiation and Plasma Medical Sciences

|October 15, 2024

Summary

This summary is machine-generated.

Deep learning models pretrained on natural images can assess [¹⁸F]FDG brain PET image quality. Unfreezing more network layers improved performance, with residual unit networks showing the best results for automated PET quality assessment.

Keywords:

Convolutional neural networks Deep learning Image quality Image reconstruction Transfer learning

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

Related Experiment Videos

Last Updated: Jun 10, 2025

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

Area of Science:

Medical Imaging
Artificial Intelligence
Radiology

Background:

Pretraining deep convolutional networks on natural images aids medical imaging tasks due to limited annotated data.
Numerous 2D pretrained backbone networks exist, but their suitability for medical image analysis varies.

Purpose of the Study:

To compare 18 different pretrained backbone networks for assessing [¹⁸F]FDG brain PET image quality.
To evaluate the impact of network over-parameterization on automated PET image quality assessment.

Main Methods:

18 ImageNet-pretrained backbones from 5 architecture groups were tested.
Networks were trained to predict clinical image quality metrics (global rating, pattern recognition, diagnostic confidence) at varying simulated PET doses.
Initial training involved only the final layer; subsequent training unfroze the last 40% of network weights.

Main Results:

Training only the final layer yielded a mean-absolute-error of ~0.5.
Unfreezing network layers reduced mean-absolute-error below 0.5 for 14 of 18 backbones.
Networks with residual units (DenseNets, ResNetV2s) generally achieved the lowest error (~0.45-0.5).

Conclusions:

Automated assessment of [¹⁸F]FDG brain PET image quality is feasible using pretrained deep learning models.
Over-parameterization, by unfreezing more network layers, can enhance performance for this task.
Residual network architectures demonstrate strong potential for robust PET image quality evaluation.