Generalizability of deep learning–based dose conversion model in proton beam therapy
Journal of Applied Clinical Medical Physics · 2026-03 · 27 (3)
DOI: 10.1002/acm2.70528 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Applied Clinical Medical Physics, online erschienen am 27.02.2026. Verfasst von Ryohei Kato und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiation Physics and Technology, Southern Tohoku Proton Therapy Center, Koriyama, Fukushima, Japan. Hauptthemen laut MEDLINE: Proton Therapy, Radiotherapy Planning, Computer-Assisted, Deep Learning, Monte Carlo Method. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Background
In proton beam therapy (PBT), the analytical pencil beam (PB) algorithm involves dose uncertainties in inhomogeneous regions, making accurate Monte Carlo (MC) dose calculation desirable but time‐consuming. Deep learning, converting the dose calculated by the PB algorithm into an MC‐equivalent dose distribution, can resolve the trade‐off between calculation accuracy and speed. Training a DL‐based dose conversion model that can be applied to any tumor site would be ideal; however, the appropriate training regions and its generalizability remain unclear.
Purpose
We developed a DL‐based dose conversion model trained on four representative tumor sites (i.e., head and neck, lung, liver, and prostate), and evaluated its generalizability.
Methods
Data from 339 patients (a total of 1147 beams) were used. PB doses were obtained from the treatment planning system, and MC doses were calculated using an in‐house MC platform. Our developed DL‐based dose conversion model was designed to input a treatment planning computed tomography image and PB dose in a single field and output an MC‐equivalent dose. The model's generalizability was evaluated on untrained tumor sites, including the esophagus, pancreas, colorectum, brain, breast, cervix, and limb bone and soft tissue. The conversion performance was assessed using 3D γ‐analysis and the Dice similarity coefficient (DSC) for isodose volumes.
Results
For most untrained tumor sites, the model achieved average γ‐passing rates of ≥90% with a criterion of 3%/2 mm. The esophagus, breast, which are close to the lung, and limb bone and soft tissue showed slightly lower passing rates of 91.3%, 85.9%, and 89.1%, respectively. The average DSC values exceeded 0.8 for most untrained tumor sites.
Conclusion
The proposed DL‐based dose conversion model demonstrated high accuracy and generalizability, even for untrained tumor sites. These findings suggest that the model can be adapted to biases in collecting disease data at each PBT center and for rare diseases.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Radiation Physics and Technology, Southern Tohoku Proton Therapy Center, Koriyama, Fukushima, Japan.
- Department of Radiation Oncology, Tohoku University Graduate School of Medicine, Sendai, Miyagi, Japan.
- Department of Radiological Sciences, School of Health Sciences, Fukushima Medical University, Fukushima, Fukushima, Japan.
- Department of Radiology, University of Yamanashi, Chuo-city, Yamanashi, Japan.
- Department of Radiation Oncology, Southern Tohoku Proton Therapy Center, Koriyama, Fukushima, Japan.
ORCID-Profile: Ryohei Kato, Noriyuki Kadoya, Takahiro Kato, Ryota Tozuka, Shuta Ogawa, Masao Murakami, Keiichi Jingu
Zitieren
APA
Kato, R., Kadoya, N., Kato, T., Tozuka, R., Ogawa, S., Murakami, M., & Jingu, K. (2026). Generalizability of deep learning–based dose conversion model in proton beam therapy. Journal of Applied Clinical Medical Physics, 27(3). https://doi.org/10.1002/acm2.70528
Vancouver
Kato R, Kadoya N, Kato T, Tozuka R, Ogawa S, Murakami M, et al. Generalizability of deep learning–based dose conversion model in proton beam therapy. Journal of Applied Clinical Medical Physics. 2026;27(3). doi:10.1002/acm2.70528
BibTeX
@article{Kato2026,
author = {Kato, Ryohei and Kadoya, Noriyuki and Kato, Takahiro and Tozuka, Ryota and Ogawa, Shuta and Murakami, Masao and Jingu, Keiichi},
title = {Generalizability of deep learning–based dose conversion model in proton beam therapy},
journal = {Journal of Applied Clinical Medical Physics},
year = {2026},
volume = {27},
number = {3},
doi = {10.1002/acm2.70528}
}
EUVIMED