EUVIMEDAtrasti EUVIMED
Open access

Generating R2 * maps from T1W and T2W images using image‐to‐image translation for Parkinson's disease

Jingzhi Wu, Chi Xiong, Xinyi Lv, Ying Yang, Wen Sun, Ying Liu, Peng Wang, Aiping Liu, Yidong Yang, Chaoshi Niu, Wei Wei, Jie Wen

Medical Physics · 2026-07 · 53 (7)

DOI: 10.1002/mp.70561 ↗

At a glance

Journal article in Medical Physics, published online on 2026-07-13. Written by Jingzhi Wu and 11 co-authors. Institution: Department of Radiology, The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology of China…. Major MEDLINE topics: Parkinson Disease, Image Processing, Computer-Assisted, Magnetic Resonance Imaging. 40 references. The full text is freely available in PubMed Central. Funded by: National Key R&D Program of China, The Chinese Academy of Science Strategic Pilot Program for Key Technology Innovation - Type C.

SantraukaSubjectsBibliografiniai duomenysCiteRelated studies

Abstract (original language)

Background

Previous studies show that quantitative R2 * mapping can reveal iron deposition and tissue alterations, potentially aiding Parkinson's disease (PD) management. However, R2 * maps are not commonly used in clinical practice due to the extra time required and sensitivity to susceptibility artifacts.

Purpose

In this work, we propose to evaluate the feasibility of using generative‐adversarial‐networks (GANs) for synthesis of R2 * maps from T1‐weighted (T1W) and T2‐weighted (T2W) images.

Methods

A GAN model was developed to synthesize R2 * maps from T1W and T2W images. 572 internal participants and 268 external participants were included. The internal‐dataset was divided into training (344), validation (114), and test data (114), while the external‐dataset was an independent test‐set. The performance of the proposed model was compared with a 2D Unet model without adversarial loss. The performance of the two models was evaluated using normalized mean square error, peak signal to noise ratio, the structure similarity index measure (SSIM), feature similarity index (FSIM), root mean square difference, average absolute difference, and relative error. Pearson method was used to assess the correlation between synthetic and real values. The area‐under‐the‐receiver‐operating‐characteristic‐curve (AUC) was calculated to evaluate the diagnostic efficacy of R2 * maps in distinguishing PD from healthy controls (HC), with a focus on the substantia nigra pars compacta (SNpc).

Results

The proposed model performed better than the 2D Unet model. In internal test‐set, high correlations were observed between synthetic and real R2 * maps across various brain regions, with coefficients ranging from 0.75 to 0.87. The AUC was 0.79 and 0.80 for synthetic and real maps ( p = 0.76), respectively. In external test‐set, the AUC was 0.84 for synthetic R2 * maps. Longitudinal analysis showed a positive correlation between ∆R2 * and ∆UPDRS (Unified‐Parkinson's‐Disease‐Rating‐Scale) in SNpc ( R = 0.69, p = 0.01) and substantia nigra pars reticulata (SNpr) ( R = 0.64, p = 0.02) for PD group.

Conclusion

The synthetic R2 * maps demonstrated good correlation with real maps and performed well in diagnosing and evaluating PD in both internal and external datasets, indicating their potential value for PD diagnosis and assessment.

Subjects

MeSH terms (MEDLINE)

Parkinson DiseaseImage Processing, Computer-AssistedMagnetic Resonance ImagingHumansGenerative Adversarial Networks

Author keywords

Parkinson's diseasegenerative adversarial networkstransverse relaxation rate constant

Authors and institutions

  • Department of Radiology, The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, Anhui, China.
  • Department of Engineering and Applied Physics, University of Science and Technology of China, Hefei, Anhui, China.
  • Core Facility Center, The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, Anhui, China.
  • Department of Neurosurgery, The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, Anhui, China.
  • Department of Neurology, The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, Anhui, China.
  • School of Information Science and Technology, University of Science and Technology of China, Hefei, Anhui, China.

Funding

  • National Key R&D Program of China
  • The Chinese Academy of Science Strategic Pilot Program for Key Technology Innovation - Type C

Cite

APA

Wu, J., Xiong, C., Lv, X., Yang, Y., Sun, W., Liu, Y., Wang, P., Liu, A., Yang, Y., Niu, C., Wei, W., & Wen, J. (2026). Generating R2 * maps from T1W and T2W images using image‐to‐image translation for Parkinson's disease. Medical Physics, 53(7). https://doi.org/10.1002/mp.70561

Vancouver

Wu J, Xiong C, Lv X, Yang Y, Sun W, Liu Y, et al. Generating R2 * maps from T1W and T2W images using image‐to‐image translation for Parkinson's disease. Medical Physics. 2026;53(7). doi:10.1002/mp.70561

BibTeX
@article{Wu2026,
  author = {Wu, Jingzhi and Xiong, Chi and Lv, Xinyi and Yang, Ying and Sun, Wen and Liu, Ying and Wang, Peng and Liu, Aiping and Yang, Yidong and Niu, Chaoshi and Wei, Wei and Wen, Jie},
  title = {Generating R2 * maps from T1W and T2W images using image‐to‐image translation for Parkinson's disease},
  journal = {Medical Physics},
  year = {2026},
  volume = {53},
  number = {7},
  doi = {10.1002/mp.70561}
}
Start for free →