EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Segmentation regularized training for multi-domain deep learning registration applied to magnetic resonance-guided prostate cancer radiotherapy

Sudharsan Madhavan, Chengcheng Gui, Lando Bosma, Josiah Simeth, Jue Jiang, Nicolas Côté, Nima Hassan Rezaeian, Himanshu Nagar, Victoria Brennan, Neelam Tyagi, Harini Veeraraghavan

Physics and Imaging in Radiation Oncology · 2026-05 · 39 · 100989

DOI: 10.1016/j.phro.2026.100989 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Physics and Imaging in Radiation Oncology, 2026-05. Verfasst von Sudharsan Madhavan und 10 weiteren Autor:innen. Beteiligte Einrichtung: Memorial Sloan Kettering Cancer Center, New York, NY, USA. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: NCI NIH HHS, NIBIB NIH HHS, National Cancer Institute Division of Cancer Epidemiology and Genetics.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background and purpose: Accurate deformable image registration (DIR) is required for contour propagation and dose accumulation for magnetic resonance guided adaptive radiotherapy (MRgART). The goal of this study was to train segmentation regularized deep learning (DL) DIR and assess domain invariant MR-MR registration. Materials and methods: DL-DIR based progressively refined registration and segmentation (ProRSeg) and VoxelMorph were trained with and without organ weighted segmentation regularization with 262 MR pairs from same domain dataset of longitudinal 3 Tesla MR simulation (MR-Sim) scans of patients with prostate cancer acquired before and every 3 months following radiotherapy. Models were compared against variational DIR by measuring accuracy of organs and clinical target volume (CTV) contour propagation accuracy on same- (58 pairs), cross- (72 1.5 Tesla MR-Linac pairs from consecutive daily treatment fractions), and mixed-domain (42 MRSim planning - MR-Linac first fraction pairs) datasets. Dose accumulation was performed for 42 patients undergoing 5-fraction MRgART on MR-Linac.

Results

ProRSeg produced robust DIR on cross- (all organs) and mixed-domain (bladder, CTV) datasets. It was the most accurate method for highly deforming organs including bladder and rectum. Segmentation regularization enhanced accuracy of both ProRSeg and VoxelMorph compared to unregularized versions. Dose accumulation feasibility study with ProRSeg indicated that 83.3% of patients met key institutional constraints for CTV coverage and bladder sparing.

Conclusions

ProRSeg was a feasible approach for multi-domain MR-MR registration for prostate cancer patients. Dose accumulation analysis indicated preliminary feasibility to evaluate compliance of delivered treatments to clinical constraints.

Schlagwörter

Stichwörter der Autor:innen

Deep learningDeformable image registrationDomain generalizationDose accumulationMR-guided radiotherapyProstate cancer

Autor:innen und Institutionen

  • Memorial Sloan Kettering Cancer Center, New York, NY, USA.
  • University Medical Center Utrecht (UMC Utrecht), Utrecht, Netherlands (the).

Förderung

  • NCI NIH HHS
  • NIBIB NIH HHS
  • National Cancer Institute Division of Cancer Epidemiology and Genetics
  • National Cancer Institute
  • National Institutes of Health
  • National Institute of Biomedical Imaging and Bioengineering

Zitieren

APA

Madhavan, S., Gui, C., Bosma, L., Simeth, J., Jiang, J., Côté, N., Rezaeian, N. H., Nagar, H., Brennan, V., Tyagi, N., & Veeraraghavan, H. (2026). Segmentation regularized training for multi-domain deep learning registration applied to magnetic resonance-guided prostate cancer radiotherapy. Physics and Imaging in Radiation Oncology, 39, 100989. https://doi.org/10.1016/j.phro.2026.100989

Vancouver

Madhavan S, Gui C, Bosma L, Simeth J, Jiang J, Côté N, et al. Segmentation regularized training for multi-domain deep learning registration applied to magnetic resonance-guided prostate cancer radiotherapy. Physics and Imaging in Radiation Oncology. 2026;39:100989. doi:10.1016/j.phro.2026.100989

BibTeX
@article{Madhavan2026,
  author = {Madhavan, Sudharsan and Gui, Chengcheng and Bosma, Lando and Simeth, Josiah and Jiang, Jue and Côté, Nicolas and Rezaeian, Nima Hassan and Nagar, Himanshu and Brennan, Victoria and Tyagi, Neelam and Veeraraghavan, Harini},
  title = {Segmentation regularized training for multi-domain deep learning registration applied to magnetic resonance-guided prostate cancer radiotherapy},
  journal = {Physics and Imaging in Radiation Oncology},
  year = {2026},
  volume = {39},
  pages = {100989},
  doi = {10.1016/j.phro.2026.100989}
}
Kostenlos starten →