Three-dimensional deep learning nomogram for post-plan quality assurance in high-dose-rate prostate brachytherapy.
- 1Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
- 2Medical Physics Unit, Department of Oncology, McGill University, Montreal, Quebec, Canada.
- 3Lady Davis Institute for Medical Research, Jewish General Hospital, Montreal, Quebec, Canada.
Abstract
Purpose
Classical nomograms for high-dose-rate (HDR) prostate brachytherapy quality assurance (QA) evaluate agreement between planned and expected total dwell time but do not directly assess the three-dimensional treated volume. This study's aim was to develop a deep learning-based 3D nomogram to predict institution-specific clinical 100% isodose volumes from anatomy and dwell positions for post-plan QA.Methods
A total of 2394 single-fraction 15 Gy HDR prostate brachytherapy plans were included. Input label maps encoded the 3D prostate, urethra, rectum, and dwell positions. The prediction target was the clinical 100% isodose volume. A five-fold cross-validation strategy was used for model development, with 2155 plans used for training and validation and 239 plans held out for testing. A 3D U-Net was trained using combined Dice and binary cross-entropy loss. Agreement was evaluated using prostate V100%, 100% isodose volume, Dice similarity coefficient, and average symmetric surface distance (ASSD). Validation-derived thresholds were used for outlier screening and targeted review.Results
The model predicted prostate V100% with a mean error of 0.73 percentage points. Predicted and clinical 100% isodose volumes showed strong agreement, with R² = 0.990, slope = 0.99, and intercept = 0.77 cm³. Median Dice similarity coefficient was 0.960, and median ASSD was 0.48 mm. A 95% threshold flagged 6.3% of plans as atypical relative to validation data.Conclusions
A deep learning-based nomogram predicted an institution-specific 100% isodose volume with low coverage error and high spatial agreement. This geometry aware post-plan QA tool can complement equation-based nomograms by directly evaluating the spatial treated volume.
Related
- Dosimetric equivalence of deep learning prostate contours after LDR brachytherapy: pre-declared margins, patient-level acceptance thresholds and the incremental predictive value of DVH indices
- Using artificial intelligence for clinical trial quality assurance in a multi-institutional neurovascular-sparing prostate radiotherapy trial
- Independent Three-dimensional Dose Calculation from Log Files for Breast VMAT: A Measurement-free Quality Assurance Approach
- A deep learning approach for EPID-based patient-specific quality assurance incorporating daily machine-state variations
- Daily Multi‐Leaf Collimator positional error prediction using machine learning: A Dynalog‐based approach for patient specific quality assurance
BibTeX
@article{beaudry2026threedimensional,
title = {{Three-dimensional deep learning nomogram for post-plan quality assurance in high-dose-rate prostate brachytherapy.}},
author = {Joel Beaudry and Tonghe Wang and Shirin A Enger and Antonio L Damato},
journal = {Physics and imaging in radiation oncology},
year = {2026},
month = sep,
doi = {10.1016/j.phro.2026.101100},
url = {https://europepmc.org/article/MED/42898876},
note = {CC BY},
}
RIS
TY - JOUR
TI - Three-dimensional deep learning nomogram for post-plan quality assurance in high-dose-rate prostate brachytherapy.
AU - Joel Beaudry
AU - Tonghe Wang
AU - Shirin A Enger
AU - Antonio L Damato
PY - 2026
DA - 2026/09/01
JO - Physics and imaging in radiation oncology
DO - 10.1016/j.phro.2026.101100
UR - https://europepmc.org/article/MED/42898876
AB - Purpose Classical nomograms for high-dose-rate (HDR) prostate brachytherapy quality assurance (QA) evaluate agreement between planned and expected total dwell time but do not directly assess the three-dimensional treated volume. This study's aim was to develop a deep learning-based 3D nomogram to predict institution-specific clinical 100% isodose volumes from anatomy and dwell positions for post-plan QA.Methods A total of 2394 single-fraction 15 Gy HDR prostate brachytherapy plans were included. Input label maps encoded the 3D prostate, urethra, rectum, and dwell positions. The prediction target was the clinical 100% isodose volume. A five-fold cross-validation strategy was used for model development, with 2155 plans used for training and validation and 239 plans held out for testing. A 3D U-Net was trained using combined Dice and binary cross-entropy loss. Agreement was evaluated using prostate V100%, 100% isodose volume, Dice similarity coefficient, and average symmetric surface distance (ASSD). Validation-derived thresholds were used for outlier screening and targeted review.Results The model predicted prostate V100% with a mean error of 0.73 percentage points. Predicted and clinical 100% isodose volumes showed strong agreement, with R² = 0.990, slope = 0.99, and intercept = 0.77 cm³. Median Dice similarity coefficient was 0.960, and median ASSD was 0.48 mm. A 95% threshold flagged 6.3% of plans as atypical relative to validation data.Conclusions A deep learning-based nomogram predicted an institution-specific 100% isodose volume with low coverage error and high spatial agreement. This geometry aware post-plan QA tool can complement equation-based nomograms by directly evaluating the spatial treated volume.
SN - 2405-6316
ER -