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

Efficient Synthesis of N-Bridged Annulenes.

Journal of the American Chemical Society·2026

Same author

Neoadjuvant retlirafusp alfa (anti-PD-L1/TGF-β bifunctional fusion protein) with or without chemotherapy in unresectable stage III non-small cell lung cancer: updated results from the phase 2 TRAILBLAZER trial.

Signal transduction and targeted therapy·2026

Same author

Multimodal voxel-wise of structural and functional brain alterations across the cerebral small vessel disease spectrum.

Neuroradiology·2026

Same author

Sintilimab plus concurrent chemoradiotherapy for treatment of locally advanced small cell lung cancer (SINCE-01): a phase II clinical trial.

Signal transduction and targeted therapy·2026

Same author

Iron-mediated reactions of gem-dihaloalkanes with α,β-unsaturated carbonyl compounds.

Chemical science·2026

Same author

Structural Volumetric Alterations in Parkinson's Disease With Mild Cognitive Impairment.

Brain and behavior·2026

Search research articles

Related Experiment Video

Updated: Jul 4, 2025

Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example

Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example

Published on: October 24, 2012

CT synthesis from MR images using frequency attention conditional generative adversarial network.

Kexin Wei¹, Weipeng Kong¹, Liheng Liu²

¹Shandong Key Laboratory of Medical Physics and Image Processing, Shandong Institute of Industrial Technology for Health Sciences and Precision Medicine, School of Physics and Electronics, Shandong Normal University, Jinan, China.

Computers in Biology and Medicine

|January 29, 2024

Summary

This study introduces a new deep learning model, FACGAN, to create synthetic CT images for MR-only radiotherapy. FACGAN generates clearer, higher-quality synthetic CT images with improved high-frequency details.

Keywords:

Attention mechanism Deep learning Generative adversarial networks MR Synthetic CT

More Related Videos

Artificial Intelligence Approaches to Assessing Primary Cilia

Artificial Intelligence Approaches to Assessing Primary Cilia

Published on: May 1, 2021

Automated Analysis of Dynamic Ca2+ Signals in Image Sequences

Automated Analysis of Dynamic Ca2+ Signals in Image Sequences

Published on: June 16, 2014

Related Experiment Videos

Last Updated: Jul 4, 2025

Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example

Mapping Cortical Dynamics Using Simultaneous MEG/EEG and Anatomically-constrained Minimum-norm Estimates: an Auditory Attention Example

Published on: October 24, 2012

Artificial Intelligence Approaches to Assessing Primary Cilia

Artificial Intelligence Approaches to Assessing Primary Cilia

Published on: May 1, 2021

Automated Analysis of Dynamic Ca2+ Signals in Image Sequences

Automated Analysis of Dynamic Ca2+ Signals in Image Sequences

Published on: June 16, 2014

Area of Science:

Medical Imaging
Radiotherapy Physics
Artificial Intelligence in Medicine

Background:

Magnetic resonance (MR) image-guided radiotherapy requires synthetic computed tomography (sCT) images for treatment planning.
Convolutional neural networks (CNNs) show promise for sCT generation but often produce blurred images with insufficient high-frequency details.

Purpose of the Study:

To develop an advanced deep learning model for generating high-quality sCT images from MR data.
To improve the accuracy and clarity of sCT images for effective radiotherapy planning.

Main Methods:

Proposed a frequency attention conditional generative adversarial network (FACGAN) incorporating a frequency cycle generative model (FCGM) and a residual frequency channel attention (RFCA) module.
Introduced high-frequency loss (HFL) and cycle consistency high-frequency loss (CHFL) for model optimization.
Validated the model on pelvic and brain datasets, comparing it with existing deep learning methods.

Main Results:

FACGAN successfully generated higher-quality sCT images compared to state-of-the-art models.
The proposed model preserved and enhanced clearer, richer high-frequency texture information in the synthetic images.
Improved inter-mapping between MR and CT data, extracting more detailed tissue structures.

Conclusions:

FACGAN effectively addresses the limitations of CNN-based sCT generation, producing superior image quality.
The enhanced high-frequency detail generation is crucial for accurate radiotherapy planning in MR-only workflows.
This approach holds significant potential for advancing MR-only radiotherapy techniques.