EUVIMEDEUVIMED entdecken
Crossref

Application of support vector machines for modeling dosimetric uncertainty in radiotherapy planning

Noah Bice, K Sunshine Osterman, Paulina Galavis, Jinyu Xue, David L Barbee, Jose R Teruel

Radiological Physics and Technology · 2026-06 · 19 (2) · 638-650

DOI: 10.1007/s12194-026-01032-2 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Radiological Physics and Technology, online erschienen am 05.03.2026. Verfasst von Noah Bice und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiation Oncology, NYU Grossman School of Medicine, NYU Langone Health, New York, NY, USA. Hauptthemen laut MEDLINE: Radiotherapy Planning, Computer-Assisted, Support Vector Machine. Die Arbeit nennt 19 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We introduce two models to describe the robustness of radiotherapy plans. First, we propose an analytical model to assess coverage probabilities as a function of target and margin size. Second, we present a novel numerical framework for the assessment of planning technique robustness. This framework is intended to be implemented at the time of commissioning a new planning technique and is not specific to a particular modality. An uncertainty-aware machine learning model is built to estimate the sensitivity of dosimetric end points to geometric errors. A Monte Carlo method is then implemented to assess the probability of meeting different coverage objectives. The numerical framework is applied to our institution’s single-isocenter multi-target cranial stereotactic radiosurgery technique. For our four-arc planning technique, our numerical model recommends a total geometric uncertainty tolerance of [Formula: see text] mm to achieve a PTV coverage [Formula: see text] with probability [Formula: see text].

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Radiotherapy Planning, Computer-AssistedSupport Vector MachineUncertaintyRadiotherapy DosageRadiometryMonte Carlo MethodRadiosurgeryHumansSoft Computing

Stichwörter der Autor:innen

Dosimetric uncertaintySingle-isocenter multi-target (SIMT)Stereotactic radiosurgery (SRS)Support vector machines (SVM)

Autor:innen und Institutionen

  • Department of Radiation Oncology, NYU Grossman School of Medicine, NYU Langone Health, New York, NY, USA.
  • Department of Radiation Oncology, Mount Sinai Hospital, New York, USA.
  • Department of Radiation Oncology, NYU Grossman School of Medicine, NYU Langone Health, New York, NY, USA. jose.teruel@nyumc.org.
  • Laura and Isaac Perlmutter Cancer Center, NYU Langone Health, 160 East 34th Street Radiation Oncology Level 1, New York, NY, 10016, USA. jose.teruel@nyumc.org.

Zitieren

APA

Bice, N., Osterman, K. S., Galavis, P., Xue, J., Barbee, D. L., & Teruel, J. R. (2026). Application of support vector machines for modeling dosimetric uncertainty in radiotherapy planning. Radiological Physics and Technology, 19(2), 638-650. https://doi.org/10.1007/s12194-026-01032-2

Vancouver

Bice N, Osterman KS, Galavis P, Xue J, Barbee DL, Teruel JR. Application of support vector machines for modeling dosimetric uncertainty in radiotherapy planning. Radiological Physics and Technology. 2026;19(2):638-650. doi:10.1007/s12194-026-01032-2

BibTeX
@article{Bice2026,
  author = {Bice, Noah and Osterman, K Sunshine and Galavis, Paulina and Xue, Jinyu and Barbee, David L and Teruel, Jose R},
  title = {Application of support vector machines for modeling dosimetric uncertainty in radiotherapy planning},
  journal = {Radiological Physics and Technology},
  year = {2026},
  volume = {19},
  number = {2},
  pages = {638--650},
  doi = {10.1007/s12194-026-01032-2}
}
Kostenlos starten →