EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep learning segmentation of the cerebral ventricular system and brainstem for pediatric radiotherapy planning

Feline Heinzelmann, Fabian Hörst, Lukas Heine, Jana Fragemann, Moritz Rempe, Gijs Luijten, Sarah Peters, Julian Hlouschek, Jan Egger, Beate Timmermann, Jens Kleesiek, Christian Bäumer

Physics and Imaging in Radiation Oncology · 2026-09 · 41 · 101086

DOI: 10.1016/j.phro.2026.101086 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Physics and Imaging in Radiation Oncology, 2026-09. Verfasst von Feline Heinzelmann und 11 weiteren Autor:innen. Beteiligte Einrichtung: West German Proton Therapy Centre Essen (WPE), Essen, Germany. Die Arbeit nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Deutsche Forschungsgemeinschaft.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background and Purpose: Segmentation of the cerebral ventricular system (VS) is central to radiotherapy, both for whole-ventricular irradiation in intracranial germ cell tumors (GCTs) and protection of the periventricular region. Complex, variable VS anatomy makes manual delineation challenging and hampers auto-segmentation model development. We reported the evaluation and workflow integration of a deep-learning (DL) model for VS and brainstem segmentation on pediatric magnetic resonance imaging (MRI). Materials and methods: An nnU-Net was trained on MRIs from 76 unique brain tumor cases (n = 73/71 for T1/T2). The evaluation used two independent retrospective cohorts: (i) GCTs evaluated on planning computed tomography (CT) images after MRI-to-CT registration mapping (n = 13/14), and (ii) brain tumors evaluated on MRI (n = 14). Performance was assessed using the Dice similarity coefficient (DSC) and Hausdorff distances (HD) against manual contours. Prospective deployment in routine clinical planning assessed usability and workflow integration.

Results

In the GCT-CT cohort, brainstem segmentations achieved median DSCs of >0.85 and HDs95 of ≤5.0/6.4 mm, while VS segmentations achieved DSCs of >0.70 and HDs95 of ≤5.3/4.3 mm, after rigid/deformable mapping to planning CT. In the MR cohort, performance on T1-/T2-weighted MRI reached DSCs of ≥0.90 for both structures, and HDs95 of ≤3.0/3.4 mm and ≤ 1.3/2.2 mm for the brainstem and VS, respectively. The DL-segmentation workflow reduced workflow time by >50% compared with manual contouring.

Conclusions

Lower performance in the GCT-CT cohort likely reflects noisier CT-based references, interobserver variability, and uncorrected geometric/anatomical mismatch. The model's accuracy supports radiotherapy planning in pediatric brain tumors through risk-structure delineation and target-volume segmentation in GCTs.

Schlagwörter

Stichwörter der Autor:innen

BrainstemDeep learning segmentationPediatric brain tumorRadiotherapy planningVentricular systemnnunet model

Autor:innen und Institutionen

  • West German Proton Therapy Centre Essen (WPE), Essen, Germany.
  • TU Dortmund University, Department of Physics, Dortmund, Germany.
  • West German Cancer Center (WTZ), University Hospital Essen, Essen, Germany.
  • Institute for Artificial Intelligence in Medicine (IKIM), University Hospital Essen, Essen, Germany.
  • University of Duisburg-Essen, Faculty of Medicine, Essen, Germany.
  • Center for Virtual and Extended Reality in Medicine (ZvRM), University Hospital Essen (AöR), Essen, Germany.

Förderung

  • Deutsche Forschungsgemeinschaft

Zitieren

APA

Heinzelmann, F., Hörst, F., Heine, L., Fragemann, J., Rempe, M., Luijten, G., Peters, S., Hlouschek, J., Egger, J., Timmermann, B., Kleesiek, J., & Bäumer, C. (2026). Deep learning segmentation of the cerebral ventricular system and brainstem for pediatric radiotherapy planning. Physics and Imaging in Radiation Oncology, 41, 101086. https://doi.org/10.1016/j.phro.2026.101086

Vancouver

Heinzelmann F, Hörst F, Heine L, Fragemann J, Rempe M, Luijten G, et al. Deep learning segmentation of the cerebral ventricular system and brainstem for pediatric radiotherapy planning. Physics and Imaging in Radiation Oncology. 2026;41:101086. doi:10.1016/j.phro.2026.101086

BibTeX
@article{Heinzelmann2026,
  author = {Heinzelmann, Feline and Hörst, Fabian and Heine, Lukas and Fragemann, Jana and Rempe, Moritz and Luijten, Gijs and Peters, Sarah and Hlouschek, Julian and Egger, Jan and Timmermann, Beate and Kleesiek, Jens and Bäumer, Christian},
  title = {Deep learning segmentation of the cerebral ventricular system and brainstem for pediatric radiotherapy planning},
  journal = {Physics and Imaging in Radiation Oncology},
  year = {2026},
  volume = {41},
  pages = {101086},
  doi = {10.1016/j.phro.2026.101086}
}
Kostenlos starten →