A two-stage framework for fast proton spot map generation in pencil beam scanning prostate SBRT planning.
- 1Department of Radiation Oncology, Mayo Clinic, Rochester, Minnesota, USA.
Abstract
Background
In pencil beam scanning (PBS) proton therapy, treatment plans are encoded as proton spot maps (PSMs). Although deep-learning methods can rapidly predict three-dimensional (3D) dose distributions from patient anatomy and planning intent, there is still no widely adopted direct, non-iterative approach to convert doses into physically deliverable spot patterns. This limits end-to-end automation in treatment planning and adaptive replanning for PBS.Purpose
We developed and retrospectively evaluated GenSpot, a two-stage framework that reconstructs a PBS PSM from CT and dose. This study focused on the dose-to-spots reconstruction step under controlled, achievable-dose conditions, using clinical Monte Carlo (MC) dose distributions as input in a single-institution prostate stereotactic body radiation therapy (SBRT) cohort.Methods
GenSpot uses a physics-informed projected proton spot map (PrPSM) representation that projects individual proton spots through the CT using water-equivalent thickness and percentage depth dose (PDD) information, aligning the spots with the CT and dose voxel grid while preserving a linear relationship with spot weights. Our dataset comprised 1036 fields from 259 prostate SBRT plans, split into training, validation, and test sets in an 80%/10%/10% ratio. A 3D SwinUNETR model was trained to predict PrPSMs from CT and dose inputs. Field-specific PSMs were then reconstructed from the predicted PrPSMs using column-wise nonnegative Lasso regression with precomputed PDD curves. MC dose distributions from GenSpot and clinical PSMs were compared using voxel-wise mean absolute error (MAE), 3D gamma analysis (3%/3 mm, 10% low-dose threshold), and dose-volume histogram (DVH) metrics at the composite-plan level.Results
Within the held-out test dataset, the selected SwinUNETR model achieved a mean PrPSM MAE of 0.06 ± 0.02 (normalized units) with high structural similarity to projected clinical PrPSMs. At the dose level, MC doses from GenSpot PSMs showed low MAE (0.07 ± 0.03 Gy in the nonzero-dose region) and high mean gamma passing rates (0.92 ± 0.05 at the field level and 0.99 ± 0.02 at the plan level). Composite-plan DVH metrics differences were generally within 1 Gy for targets and organs at risk, although the high-dose tail of the clinical target volume showed a modest systematic increase. Spot map complexity, measured by the number of energy layers and spot counts, was similar to that of clinical plans, with 20% more spots for GenSpot PSMs. For a four-field case, GenSpot PSM generation required approximately 2.9 ± 0.2 min excluding final MC recalculation.Conclusions
GenSpot reconstructed machine-compatible PSMs from CT and clinical MC dose with close composite-plan dose agreement in a single-institution prostate SBRT cohort. These results support GenSpot as a rapid, physics-informed dose-to-spots reconstruction component under achievable-dose conditions. The present study does not establish performance for arbitrary predicted or adapted dose inputs, and final MC recalculation and standard QA remain required before clinical use. Further validation with non-idealized dose inputs, multi-institution data, higher-resolution dose representations, and dedicated optimization-based baselines is needed before broader clinical implementation.
Related
- An operational key performance indicator framework and reference data from a multi-room pencil beam scanning proton therapy center
- Radiation-induced tumor following pencil beam scanning proton therapy for pediatric intracranial ependymoma: a pragmatic analysis of risk factors
- A Comprehensive Beam-Level Assessment of Pencil-Beam Scanning Migration and Utilization at a Multi-System Proton Therapy Center.
- Design, implementation, and assessment of an independent second-check system for proton and carbon-ion pencil-beam scanning: delivery-aware Monte Carlo and microdosimetric RBE-weighted dose
- AI in Proton Therapy Treatment Planning: A Narrative Review.
BibTeX
@article{tang2026twostage,
title = {{A two-stage framework for fast proton spot map generation in pencil beam scanning prostate SBRT planning.}},
author = {Xueyan Tang and Hok Wan Chan Tseung and Mark Pepin and Jiasen Ma and David M Routman and Doug J Moseley and Brandon Reber and Jed E Johnson and Jing Qian},
journal = {Medical physics},
year = {2026},
month = oct,
doi = {10.1002/mp.70693},
url = {https://europepmc.org/article/MED/42852689},
note = {CC BY-NC},
}
RIS
TY - JOUR
TI - A two-stage framework for fast proton spot map generation in pencil beam scanning prostate SBRT planning.
AU - Xueyan Tang
AU - Hok Wan Chan Tseung
AU - Mark Pepin
AU - Jiasen Ma
AU - David M Routman
AU - Doug J Moseley
AU - Brandon Reber
AU - Jed E Johnson
AU - Jing Qian
PY - 2026
DA - 2026/10/01
JO - Medical physics
DO - 10.1002/mp.70693
UR - https://europepmc.org/article/MED/42852689
AB - Background In pencil beam scanning (PBS) proton therapy, treatment plans are encoded as proton spot maps (PSMs). Although deep-learning methods can rapidly predict three-dimensional (3D) dose distributions from patient anatomy and planning intent, there is still no widely adopted direct, non-iterative approach to convert doses into physically deliverable spot patterns. This limits end-to-end automation in treatment planning and adaptive replanning for PBS.Purpose We developed and retrospectively evaluated GenSpot, a two-stage framework that reconstructs a PBS PSM from CT and dose. This study focused on the dose-to-spots reconstruction step under controlled, achievable-dose conditions, using clinical Monte Carlo (MC) dose distributions as input in a single-institution prostate stereotactic body radiation therapy (SBRT) cohort.Methods GenSpot uses a physics-informed projected proton spot map (PrPSM) representation that projects individual proton spots through the CT using water-equivalent thickness and percentage depth dose (PDD) information, aligning the spots with the CT and dose voxel grid while preserving a linear relationship with spot weights. Our dataset comprised 1036 fields from 259 prostate SBRT plans, split into training, validation, and test sets in an 80%/10%/10% ratio. A 3D SwinUNETR model was trained to predict PrPSMs from CT and dose inputs. Field-specific PSMs were then reconstructed from the predicted PrPSMs using column-wise nonnegative Lasso regression with precomputed PDD curves. MC dose distributions from GenSpot and clinical PSMs were compared using voxel-wise mean absolute error (MAE), 3D gamma analysis (3%/3 mm, 10% low-dose threshold), and dose-volume histogram (DVH) metrics at the composite-plan level.Results Within the held-out test dataset, the selected SwinUNETR model achieved a mean PrPSM MAE of 0.06 ± 0.02 (normalized units) with high structural similarity to projected clinical PrPSMs. At the dose level, MC doses from GenSpot PSMs showed low MAE (0.07 ± 0.03 Gy in the nonzero-dose region) and high mean gamma passing rates (0.92 ± 0.05 at the field level and 0.99 ± 0.02 at the plan level). Composite-plan DVH metrics differences were generally within 1 Gy for targets and organs at risk, although the high-dose tail of the clinical target volume showed a modest systematic increase. Spot map complexity, measured by the number of energy layers and spot counts, was similar to that of clinical plans, with 20% more spots for GenSpot PSMs. For a four-field case, GenSpot PSM generation required approximately 2.9 ± 0.2 min excluding final MC recalculation.Conclusions GenSpot reconstructed machine-compatible PSMs from CT and clinical MC dose with close composite-plan dose agreement in a single-institution prostate SBRT cohort. These results support GenSpot as a rapid, physics-informed dose-to-spots reconstruction component under achievable-dose conditions. The present study does not establish performance for arbitrary predicted or adapted dose inputs, and final MC recalculation and standard QA remain required before clinical use. Further validation with non-idealized dose inputs, multi-institution data, higher-resolution dose representations, and dedicated optimization-based baselines is needed before broader clinical implementation.
SN - 0094-2405
SN - 2473-4209
ER -