EUVIMEDEUVIMED entdecken
Open Access

Meta-Learning Initializations for Interactive Medical Image Registration

Zachary M C Baum, Yipeng Hu, Dean C Barratt

IEEE Transactions on Medical Imaging · 2023-03 · 42 (3) · 823-833

DOI: 10.1109/tmi.2022.3218147 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Medical Imaging, 2023-03. Verfasst von Zachary M C Baum und 2 weiteren Autor:innen. Hauptthemen laut MEDLINE: Prostate, Imaging, Three-Dimensional. Die Arbeit wurde bisher 21-mal zitiert (Crossref) und nennt 71 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: Wellcome Trust, Cancer Research UK, Natural Sciences and Engineering Research Council of Canada Postgraduate Scholarships-Doctoral Program.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We present a meta-learning framework for interactive medical image registration. Our proposed framework comprises three components: a learning-based medical image registration algorithm, a form of user interaction that refines registration at inference, and a meta-learning protocol that learns a rapidly adaptable network initialization. This paper describes a specific algorithm that implements the registration, interaction and meta-learning protocol for our exemplar clinical application: registration of magnetic resonance (MR) imaging to interactively acquired, sparsely-sampled transrectal ultrasound (TRUS) images. Our approach obtains comparable registration error (4.26 mm) to the best-performing non-interactive learning-based 3D-to-3D method (3.97 mm) while requiring only a fraction of the data, and occurring in real-time during acquisition. Applying sparsely sampled data to non-interactive methods yields higher registration errors (6.26 mm), demonstrating the effectiveness of interactive MR-TRUS registration, which may be applied intraoperatively given the real-time nature of the adaptation process.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ProstateImaging, Three-DimensionalMaleHumansUltrasonographyAlgorithmsMagnetic Resonance Imaging

Förderung

  • Wellcome Trust
  • Cancer Research UK
  • Natural Sciences and Engineering Research Council of Canada Postgraduate Scholarships-Doctoral Program
  • University College London Overseas and Graduate Research Scholarships
  • International Alliance for Cancer Early Detection, Alliance Between Cancer Research U.K
  • Canary Center at Stanford University
  • University of Cambridge
  • Knight Cancer Institute, Oregon Health and Science University

Zitieren

APA

Baum, Z. M. C., Hu, Y., & Barratt, D. C. (2023). Meta-Learning Initializations for Interactive Medical Image Registration. IEEE Transactions on Medical Imaging, 42(3), 823-833. https://doi.org/10.1109/tmi.2022.3218147

Vancouver

Baum ZMC, Hu Y, Barratt DC. Meta-Learning Initializations for Interactive Medical Image Registration. IEEE Transactions on Medical Imaging. 2023;42(3):823-833. doi:10.1109/tmi.2022.3218147

BibTeX
@article{Baum2023,
  author = {Baum, Zachary M C and Hu, Yipeng and Barratt, Dean C},
  title = {Meta-Learning Initializations for Interactive Medical Image Registration},
  journal = {IEEE Transactions on Medical Imaging},
  year = {2023},
  volume = {42},
  number = {3},
  pages = {823--833},
  doi = {10.1109/tmi.2022.3218147}
}
Kostenlos starten →