EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

An effective deep learning algorithm for medical image registration

Jinqiu Deng iD, Ke Chen iD, Mingke Li, Zhichao Zuo, Alejandro F Frangi, Jianping Zhang iD

PLOS Digital Health · 16.04.2026 · 5 (4) · e0001339

DOI: 10.1371/journal.pdig.0001339 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS Digital Health, online erschienen am 16.04.2026. Verfasst von Jinqiu Deng und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Mathematics and Computational Science, Xiangtan University, Xiangtan, China. Die Arbeit nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Science and technology innovation Program of Hunan Province, Natural Science Foundation of Hunan Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Image registration is crucial for many medical imaging applications, including longitudinal monitoring and multimodal information fusion. A key challenge is to achieve accurate alignment while strictly preserving topology and invertibility. To address the limitations of traditional penalty-based regularization, which may still permit local folding, this study proposes DTC-Reg, a dynamically learned registration framework that more explicitly enforces diffeomorphic deformation. The framework integrates a homotopy-based control–increment formulation with explicit multiscale geometric constraints. Two parameter-sharing U-Nets first extract multiscale feature pyramids from the input images, after which a symmetric registration module with a sequential temporal cascade network progressively refines the forward and inverse multiscale deformation fields. To further enhance diffeomorphic consistency, this study introduces a Multiscale Folding-aware Deformation Correction (MFDC) module that explicitly detects and geometrically rectifies folding points in the predicted deformation fields. Beyond its integration within DTC-Reg, MFDC can also be readily incorporated into several state-of-the-art registration networks, significantly reducing folding and improving deformation regularity. Extensive experiments on three 3D brain MRI registration tasks demonstrate that the proposed method consistently achieves superior performance over existing approaches in both quantitative and qualitative evaluations.

Autor:innen und Institutionen

  • School of Mathematics and Computational Science, Xiangtan University, Xiangtan, China.
  • Department of Mathematics and Statistics, University of Strathclyde, Glasgow, United Kingdom.
  • Radiology Department of Xiangtan Central Hospital, Xiangtan, China.
  • Schools of Health Sciences and Computer Science, University of Manchester, Manchester, United Kingdom.

ORCID-Profile: Jinqiu Deng, Ke Chen, Jianping Zhang

Förderung

  • National Natural Science Foundation of China
  • Science and technology innovation Program of Hunan Province
  • Natural Science Foundation of Hunan Province
  • Royal Academy of Engineering Chair INSILEX
  • UKRI Frontier Research Guarantee INSILICO

Zitieren

APA

Deng, J., Chen, K., Li, M., Zuo, Z., Frangi, A. F., & Zhang, J. (2026). An effective deep learning algorithm for medical image registration. PLOS Digital Health, 5(4), e0001339. https://doi.org/10.1371/journal.pdig.0001339

Vancouver

Deng J, Chen K, Li M, Zuo Z, Frangi AF, Zhang J. An effective deep learning algorithm for medical image registration. PLOS Digital Health. 2026;5(4):e0001339. doi:10.1371/journal.pdig.0001339

BibTeX
@article{Deng2026,
  author = {Deng, Jinqiu and Chen, Ke and Li, Mingke and Zuo, Zhichao and Frangi, Alejandro F and Zhang, Jianping},
  title = {An effective deep learning algorithm for medical image registration},
  journal = {PLOS Digital Health},
  year = {2026},
  volume = {5},
  number = {4},
  pages = {e0001339},
  doi = {10.1371/journal.pdig.0001339}
}
Kostenlos starten →