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
Abstract
Background and purpose: Dose Volume Histogram (DVH) indices remain the main dose-effect metrics for toxicity prediction, but they depend on how contours were made. Inter-observer variability (IOV) is unavoidable and clinically accepted, so the question for automatic segmentation addresses equivalence: do automatic contours produce DVH errors comparable to human IOV, and can these indices predict patient-reported toxicity? Materials and methods: In 429 patients treated with iodine-125 LDR-BT monotherapy, indices from expert manual delineation were compared with a deterministic and a Bayesian nnUNet on a fixed dose distribution, by two-one-sided tests against margins set from CT contouring IOV. Logistic regression gave patient-level thresholds at 90\,\% probability of equivalence. In 380 patients, eleven DHV indices were added to a clinical baseline predicting change in International Prostate Symptom Score (IPSS) at six horizons over 5 years, with nested cross-validation, bootstrap intervals and corrected t-tests. Results: All cohort-level comparisons of DVH indices were declared equivalent, with no interval consuming more than 44\,\% of its equivalence margin. Individual agreement was weaker with equivalence rates from 55.9\,\% to 89.7\,\%, depending on the DVH index and automatic segmentation model. Thresholds ranged from 0.864 to 0.958 Dice. No DVH block improved IPSS prediction at any horizon. The largest improvement declared by bootstrap intervals was 0.12 IPSS points, well below the minimal clinically important difference, across all learners. Conclusion: Automatic contours matched expert dosimetry within human IOV on average, but individual equivalence requires a volume dependent quality metric threshold definition. Our DVH indices panel provides no significant predictive power for IPSS, irrespective of segmentation source and horizon.
Related
- Dosimetric and Treatment Delivery Efficiency Comparison of Two Ring-gantry Linear Accelerator Platforms for Liver Stereotactic Body Radiotherapy
- In vivo Dose Measurement in Cervical Cancer Brachytherapy Using Dysprosium-doped Calcium Sulfate Thermoluminescent Dosimeters: A Phantom Study
- Three-dimensional deep learning nomogram for post-plan quality assurance in high-dose-rate prostate brachytherapy.
- Dosimetric Comparison of Functional Radiosurgery Treatments for Trigeminal Neuralgia, Essential Tremor, and Obsessive-compulsive Disorder Using Gamma Knife, CyberKnife, and Linear Accelerators
- CBCT-Guided Online Adaptive Ultrahypofractionated Whole-Breast Radiotherapy for Early-Stage Breast Cancer: Dosimetric Outcomes and Acute Toxicity.
BibTeX
@misc{stcyr2026dosimetric,
title = {{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}},
author = {Louis-Bernard St-Cyr and Anne Saint-Laurent and José Angel Lesteiro-Tejeda and Sylviane Aubin and Marie-Claude Lavallée and Luc Beaulieu and Éric Vigneault and André-Guy Martin and François-Olivier Fabi and Louis Archambault},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.14028},
eprint = {2609.14028},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.14028},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - UNPB
TI - 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
AU - Louis-Bernard St-Cyr
AU - Anne Saint-Laurent
AU - José Angel Lesteiro-Tejeda
AU - Sylviane Aubin
AU - Marie-Claude Lavallée
AU - Luc Beaulieu
AU - Éric Vigneault
AU - André-Guy Martin
AU - François-Olivier Fabi
AU - Louis Archambault
PY - 2026
DA - 2026/09/12
PB - arXiv
DO - 10.48550/arxiv.2609.14028
UR - https://arxiv.org/abs/2609.14028
AB - Background and purpose: Dose Volume Histogram (DVH) indices remain the main dose-effect metrics for toxicity prediction, but they depend on how contours were made. Inter-observer variability (IOV) is unavoidable and clinically accepted, so the question for automatic segmentation addresses equivalence: do automatic contours produce DVH errors comparable to human IOV, and can these indices predict patient-reported toxicity? Materials and methods: In 429 patients treated with iodine-125 LDR-BT monotherapy, indices from expert manual delineation were compared with a deterministic and a Bayesian nnUNet on a fixed dose distribution, by two-one-sided tests against margins set from CT contouring IOV. Logistic regression gave patient-level thresholds at 90\,\% probability of equivalence. In 380 patients, eleven DHV indices were added to a clinical baseline predicting change in International Prostate Symptom Score (IPSS) at six horizons over 5 years, with nested cross-validation, bootstrap intervals and corrected t-tests. Results: All cohort-level comparisons of DVH indices were declared equivalent, with no interval consuming more than 44\,\% of its equivalence margin. Individual agreement was weaker with equivalence rates from 55.9\,\% to 89.7\,\%, depending on the DVH index and automatic segmentation model. Thresholds ranged from 0.864 to 0.958 Dice. No DVH block improved IPSS prediction at any horizon. The largest improvement declared by bootstrap intervals was 0.12 IPSS points, well below the minimal clinically important difference, across all learners. Conclusion: Automatic contours matched expert dosimetry within human IOV on average, but individual equivalence requires a volume dependent quality metric threshold definition. Our DVH indices panel provides no significant predictive power for IPSS, irrespective of segmentation source and horizon.
ER -