• Home
  • Search
  • Image enhancement for accelerated MRI using a joint GAN and diffusion model framework.
  • Cite Icon1
  • https://doi.org/10.1002/mp.70242Copy DOI Icon

Image enhancement for accelerated MRI using a joint GAN and diffusion model framework.

Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Magnetic resonance imaging (MRI) is a defining feature of magnetic resonance image-guided radiotherapy (MRigRT), providing superior soft-tissue contrast compared to conventional cone beam CT (CBCT). However, conventional MRI protocols typically require prolonged acquisition times. High-speed scanning techniques implemented to improve clinical workflow efficiency often produce low-quality images, characterized by reduced resolution and blurring, which compromise the accuracy of downstream tasks. Although deep learning-based enhancement methods have shown promise, most fail to adequately correct residual errors between predicted and actual images, introducing uncertainties in clinicaldecisions. This study aims to propose and evaluate a deep learning-based approach to enhance the image quality of accelerated MRI acquisitions, with the dual objectives of ensuring registration accuracy for precise tumor targeting, and shortening in-room time to minimize patient discomfort on the treatmentcouch. We acquired 72 paired 3D T2-weighted MRI scans from 62 glioma patients using a 1.5T Unity MR-Linac, with each pair comprising a standard protocol scan and an accelerated protocol scan. We proposed Residual Refinement Enhancement Network (RRENet), a novel end-to-end deep learning framework that combines generative adversarial network (GAN) and diffusion model (DM) for accelerated MRI enhancement. The framework operates in a two-stage process: first, a GAN-based predictor generates the low-frequency content; subsequently, a DM-based module refines the output by estimating residual errors. To further improve performance, an additional High-frequency Separation Training Module (HSTM) was incorporated to preserve fine anatomical details. Image quality was assessed using peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and root mean square error (RMSE). Clinical applicability was evaluated through rigid registration analysis with commercialsoftware. RRENet demonstrated superior performance compared to competing methods, achieving a 70% reduction in imaging time (vs. standard protocol) while maintaining high quality (SSIM = 92.27 2.03%, PSNR = 32.84 1.54dB, RMSE = 0.025). Enhanced images retained fine anatomical structures alignment ( 2.2 mm translation, 0.2 rotation) to planning CT, ensuring the accuracy of patient setupcorrection. The proposed method can generate high-quality images from high-speed scanning sequences, reducing acquisition time while ensuring precise tumor targeting through reliable imageregistration.

Similar Papers
  • Research Article
  • Citations2

Patient- and fraction-specific magnetic resonance volume reconstruction from orthogonal images with generative adversarial networks.

  • Feb 04, 2025
  • Medical physics
  • Hideaki Hirashima +9
  • Research Article

A bias field correction workflow based on generative adversarial network for abdominal cancers treated with 0.35T MR-LINAC.

  • Jan 01, 2026
  • Journal of applied clinical medical physics
  • Ching-Ching Yang +1
  • Research Article
  • Citations9

Utilizing deep learning techniques to improve image quality and noise reduction in preclinical low-dose PET images in the sinogram domain.

  • Nov 15, 2023
  • Medical Physics
  • Kishore Krishnagiri Manoj Doss +1
  • Research Article
  • Citations29

Detecting brain lesions in suspected acute ischemic stroke with CT-based synthetic MRI using generative adversarial networks

  • Jan 01, 2022
  • Annals of Translational Medicine
  • Na Hu +9
  • Research Article
  • Citations16

Refinement of image quality in panoramic radiography using a generative adversarial network.

  • May 02, 2023
  • Dento maxillo facial radiology
  • Hak-Sun Kim +6
  • Research Article
  • Citations133

Improving CBCT quality to CT level using deep learning with generative adversarial network.

  • May 14, 2021
  • Medical physics
  • Yang Zhang +8
  • Research Article

BrainGAN Framework for Generating Synthetic Contrast-Enhanced Multisequence MRI in Brain Metastases.

  • Jul 09, 2026
  • Journal of imaging informatics in medicine
  • Merve Solak +5
  • Research Article
  • Citations7

Clinical ultra-high resolution CT scans enabled by using a generative adversarial network.

  • Jan 11, 2023
  • Medical Physics
  • Yao Sun +11
  • Research Article
  • Citations12

Bridging the Camera Domain Gap With Image-to-Image Translation Improves Glaucoma Diagnosis.

  • Dec 22, 2023
  • Translational Vision Science & Technology
  • Shuang He +9
  • Research Article

Metal artifact reduction and clinical verification in oral and maxillofacial region based on deep learning

  • Jun 09, 2023
  • Zhonghua kou qiang yi xue za zhi = Zhonghua kouqiang yixue zazhi = Chinese journal of stomatology
  • W Zeng +3
  • Research Article

Generative Synthesis of Fractional Anisotropy Maps from T1 MRI Using Transfer Learning for White Matter Assessment in Stroke.

  • Apr 24, 2026
  • Brain topography
  • Gyubin Kwon +3
  • Research Article

Synthesizing cone-beam projections in radiotherapy using deep learning network for patients with head and neck cancer.

  • Jul 01, 2026
  • Quantitative imaging in medicine and surgery
  • Yuhan Fan +8
  • Research Article

AI-Based RGB Image Generation From RG Fundus Images Using Pix2Pix: Validation by Quantitative and Observer-Based Evaluations

  • Jan 09, 2026
  • Translational Vision Science & Technology
  • Kumiko Kato +9
  • Research Article

Advanced deep learning for PET image reconstruction and quality improvement

  • Oct 04, 2025
  • The Imaging Science Journal
  • Asma Benyelles +1
  • Research Article
  • Citations13

A geometry-guided deep learning technique for CBCT reconstruction

  • Jul 30, 2021
  • Physics in Medicine & Biology
  • Ke Lu +2
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.