EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Personalized deep learning auto‐segmentation models for adaptive fractionated magnetic resonance‐guided radiation therapy of the abdomen

Maria Kawula, Sebastian Marschner, Chengtao Wei, Marvin F Ribeiro, Stefanie Corradini, Claus Belka, Guillaume Landry, Christopher Kurz

Medical Physics · 2025

DOI: 10.1002/mp.17580 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Medical Physics, 2025. Verfasst von Maria Kawula und 7 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiation Oncology, LMU University Hospital, LMU Munich, Munich, Germany. Hauptthemen laut MEDLINE: Deep Learning, Radiotherapy, Image-Guided, Magnetic Resonance Imaging, Dose Fractionation, Radiation. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Wilhelm Sander-Stiftung.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Manual contour corrections during fractionated magnetic resonance (MR)‐guided radiotherapy (MRgRT) are time‐consuming. Conventional population models for deep learning auto‐segmentation might be suboptimal for MRgRT at MR‐Linacs since they do not incorporate manual segmentation from treatment planning and previous fractions.

Purpose

In this work, we investigate patient‐specific (PS) auto‐segmentation methods leveraging expert‐segmented planning and prior fraction MR images (MRIs) to improve auto‐segmentation on consecutive treatment days. Materials and Methods Data from 151 abdominal cancer patients treated at a 0.35 T MR‐Linac (151 planning and 215 fraction MRIs) were included. Population baseline models (BMs) were trained on 107 planning MRIs for one‐class segmentation of the aorta, bowel, duodenum, kidneys, liver, spinal canal, and stomach. PS models were obtained by fine‐tuning the BMs using the planning MRI (). Maximal improvement by continuously updating the PS models was investigated by adding the first four out of five fraction MRIs (). Similarly, PS models without BM were trained ( and ). All hyperparameters were optimized using 23 patients, and the methods were tested on the remaining 21 patients. Evaluation involved Dice similarity coefficient (DSC), average () and the 95 th percentile (HD 95 ) Hausdorff distance. A qualitative contour assessment by a radiation oncologist was performed for BM, , and . Results and networks had the best geometric performance. and BMs showed similar DSC and HDs values, however models outperformed BMs. predictions scored the best in the qualitative evaluation, followed by the BMs and models.

Conclusion

Personalized auto‐segmentation models outperformed the population BMs. In most cases, delineations were judged to be directly usable for treatment adaptation without further corrections, suggesting a potential time saving during fractionated treatment.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningRadiotherapy, Image-GuidedMagnetic Resonance ImagingDose Fractionation, RadiationAbdomenImage Processing, Computer-AssistedPrecision MedicineAbdominal NeoplasmsHumansRadiotherapy Planning, Computer-AssistedAutomationMale

Stichwörter der Autor:innen

MR‐Linacauto‐segmentationpatient‐specific transfer learning

Autor:innen und Institutionen

  • Department of Radiation Oncology, LMU University Hospital, LMU Munich, Munich, Germany.
  • German Cancer Consortium (DKTK), partner site Munich, a partnership between DKFZ and LMU University Hospital Munich, Munich, Germany.
  • Bavarian Cancer Research Center (BZKF), Munich, Germany.

Förderung

  • Wilhelm Sander-Stiftung

Zitieren

APA

Kawula, M., Marschner, S., Wei, C., Ribeiro, M. F., Corradini, S., Belka, C., Landry, G., & Kurz, C. (2025). Personalized deep learning auto‐segmentation models for adaptive fractionated magnetic resonance‐guided radiation therapy of the abdomen. Medical Physics. https://doi.org/10.1002/mp.17580

Vancouver

Kawula M, Marschner S, Wei C, Ribeiro MF, Corradini S, Belka C, et al. Personalized deep learning auto‐segmentation models for adaptive fractionated magnetic resonance‐guided radiation therapy of the abdomen. Medical Physics. 2025. doi:10.1002/mp.17580

BibTeX
@article{Kawula2025,
  author = {Kawula, Maria and Marschner, Sebastian and Wei, Chengtao and Ribeiro, Marvin F and Corradini, Stefanie and Belka, Claus and Landry, Guillaume and Kurz, Christopher},
  title = {Personalized deep learning auto‐segmentation models for adaptive fractionated magnetic resonance‐guided radiation therapy of the abdomen},
  journal = {Medical Physics},
  year = {2025},
  doi = {10.1002/mp.17580}
}
Kostenlos starten →