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

Synthetic CT generation from cone-beam CT using deep learning for head-and-neck adaptive radiotherapy

Marcos M Sánchez, Carlos F Gracia, Concepción H Martínez, Moisés S Beltrán, David G Riñon, Giorgia Yang, Raúl S López, Pablo G Fernández, Iván A Delgado, Daniel R Navarrete, Miguel B Miranda, Pedro P Marcos und 3 weitere

Physics and Imaging in Radiation Oncology · 2026-07 · 40 · 101033

DOI: 10.1016/j.phro.2026.101033 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Physics and Imaging in Radiation Oncology, 2026-07. Verfasst von Marcos M Sánchez und 14 weiteren Autor:innen. Beteiligte Einrichtung: Hospital Universitario La Paz, Madrid, Spain. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 68 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background and purpose: Synthetic computed tomography generation from cone-beam computed tomography enables adaptive radiotherapy on conventional linear accelerators, but its clinical implementation is limited. We developed and validated a deep learning model for synthetic computed tomography generation in head-and-neck patients, including regions beyond the cone-beam computed tomography field of view, and evaluated its accuracy for adaptive radiotherapy workflows. Materials and methods: A UNet3+ generative adversarial network with multi-component losses was trained on 148 patients (90 training, 38 validation, 20 independent testing). Anatomical concordance, intensity fidelity versus the deformably registered planning computed tomography, and dose recalculation in a commercial treatment planning system (global gamma at 3%/2 mm and 2%/2 mm, 10% dose threshold) were assessed.

Results

In Hounsfield units, mean absolute error was 49.9 ± 11.0 globally, 40.2 ± 8.2 inside the cone-beam field of view (mean error 0.2 ± 2.7) and 61.4 ± 19.0 outside. Mean differences in planning target volume dose metrics were < 1.6% (maximum 5.1%), and mean gamma pass rates were 99.0% (3%, 2 mm) and 98.7% (2%, 2 mm). Surface Dice at a 2-mm tolerance was 0.81 ± 0.06 (Dice 0.97 ± 0.01; mean surface distance 1.07 ± 0.31 mm; volume similarity 0.99 ± 0.01).

Conclusions

The proposed model generated accurate, anatomically coherent synthetic computed tomography from cone-beam computed tomography, including realistic reconstruction beyond the cone-beam field of view, supporting the feasibility for daily adaptive-radiotherapy dose recalculation on conventional linear accelerators and motivating prospective clinical evaluation.

Schlagwörter

Stichwörter der Autor:innen

Synthetic CTadaptive radiotherapycone-beam CTdose recalculationgenerative adversarial network

Autor:innen und Institutionen

  • Hospital Universitario La Paz, Madrid, Spain.

Zitieren

APA

Sánchez, M. M., Gracia, C. F., Martínez, C. H., Beltrán, M. S., Riñon, D. G., Yang, G., López, R. S., Fernández, P. G., Delgado, I. A., Navarrete, D. R., Miranda, M. B., Marcos, P. P., Robles, R. G., De Los Ríos, A. L., & Roldán, M. R. (2026). Synthetic CT generation from cone-beam CT using deep learning for head-and-neck adaptive radiotherapy. Physics and Imaging in Radiation Oncology, 40, 101033. https://doi.org/10.1016/j.phro.2026.101033

Vancouver

Sánchez MM, Gracia CF, Martínez CH, Beltrán MS, Riñon DG, Yang G, et al. Synthetic CT generation from cone-beam CT using deep learning for head-and-neck adaptive radiotherapy. Physics and Imaging in Radiation Oncology. 2026;40:101033. doi:10.1016/j.phro.2026.101033

BibTeX
@article{Snchez2026,
  author = {Sánchez, Marcos M and Gracia, Carlos F and Martínez, Concepción H and Beltrán, Moisés S and Riñon, David G and Yang, Giorgia and López, Raúl S and Fernández, Pablo G and Delgado, Iván A and Navarrete, Daniel R and Miranda, Miguel B and Marcos, Pedro P and Robles, Rocio G and De Los Ríos, Alonso L and Roldán, Marta R},
  title = {Synthetic CT generation from cone-beam CT using deep learning for head-and-neck adaptive radiotherapy},
  journal = {Physics and Imaging in Radiation Oncology},
  year = {2026},
  volume = {40},
  pages = {101033},
  doi = {10.1016/j.phro.2026.101033}
}
Kostenlos starten →