Francisco Berumen
Research Subjects
Research Interests
Francisco Berumen has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Luc Beaulieu
Physique, génie physique et optique, Universite Laval, 2375 rue de la Terrasse, 1401 18e rue, Quebec, G1J 1Z4, CANADA
5 shared publications Joseph Perl
SLAC National Accelerator Laboratory, Menlo Park, CA, United States of America
2 shared publications- Samuel Ouellet
Department of Physics, Carleton Laboratory for Radiotherapy Physics, Carleton University, Ottawa, Ontario, Canada; Département de physique, de génie physique et d'optique et Centre de recherche sur le cancer, Université Laval, Québec, Canada; Service de radio-oncologie et Axe Oncologie du CRCHU de Québec, CHU de Québec - Université Laval, Québec, Canada
2 shared publications - Shirin Abbasinejad Enger
McGill University
2 shared publications Yunzhi Ma
Service de physique médicale et de radioprotection, Centre Intégré de Cancérologie, CHU de Québec - Université Laval et Centre de recherche du CHU de Québec, Québec, Québec, Canada; Département de Physique, de Génie Physique et d'Optique et Centre de Recherche sur le Cancer, Université Laval, Québec Québec G1V 0A6, Canada
2 shared publicationsAndré-Guy Martin
Service de radio-oncologie, CHU de Québec-Université Laval, 2250 Henri-Bourassa, Quebec, Quebec, G1G 5X1, CANADA
1 shared publication
Latest Publications
- 9 April 2024
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 69 (8) , 085026
Aleatoric and epistemic uncertainty extraction of patient-specific deep learning-based dose predictions in LDR prostate brachytherapy
Objective. In brachytherapy, deep learning (DL) algorithms have shown the capability of predicting 3D dose volumes. The reliability and accuracy of such methodologies remain under scrutiny for prospective clinical applications. This study aims to establish fast DL-based predictive dose algorithms for low-dose rate (LDR) prostate brachytherapy and to evaluate their uncertainty and stability. Approach. Data from 200 prostate patients, treated with 125 I sources, was collected. The Monte Carlo (MC) ground truth dose volumes were calculated with TOPAS considering the interseed effects and an organ-based material assignment. Two 3D convolutional neural networks, UNet and ResUNet TSE, were trained using the patient geometry and the seed positions as the input data. The dataset was randomly split into training (150), validation (25) and test (25) sets. The aleatoric (associated with the input data) and epistemic (associated with the model) uncertainties of the DL models were assessed. Main results. For the full test set, with respect to the MC reference, the predicted prostate D 90 metric had mean differences of −0.64% and 0.08% for the UNet and ResUNet TSE models, respectively. In voxel-by-voxel comparisons, the average global dose difference ratio in the [−1%, 1%] range included 91.0% and 93.0% of voxels for the UNet and the ResUNet TSE, respectively. One forward pass or prediction took 4 ms for a 3D dose volume of 2.56 M voxels (128 × 160 × 128). The ResUNet TSE model closely encoded the well-known physics of the problem as seen in a set of uncertainty maps. The ResUNet TSE rectum D 2cc had the largest uncertainty metric of 0.0042. Significance. The proposed DL models serve as rapid dose predictors that consider the patient anatomy and interseed attenuation effects. The derived uncertainty is interpretable, highlighting areas where DL models may struggle to provide accurate estimations. The uncertainty analysis offers a comprehensive evaluation tool for dose predictor model assessment.
- 9 April 2024
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 69 (8) , 085026
Aleatoric and epistemic uncertainty extraction of patient-specific deep learning-based dose predictions in LDR prostate brachytherapy
Objective. In brachytherapy, deep learning (DL) algorithms have shown the capability of predicting 3D dose volumes. The reliability and accuracy of such methodologies remain under scrutiny for prospective clinical applications. This study aims to establish fast DL-based predictive dose algorithms for low-dose rate (LDR) prostate brachytherapy and to evaluate their uncertainty and stability. Approach. Data from 200 prostate patients, treated with 125 I sources, was collected. The Monte Carlo (MC) ground truth dose volumes were calculated with TOPAS considering the interseed effects and an organ-based material assignment. Two 3D convolutional neural networks, UNet and ResUNet TSE, were trained using the patient geometry and the seed positions as the input data. The dataset was randomly split into training (150), validation (25) and test (25) sets. The aleatoric (associated with the input data) and epistemic (associated with the model) uncertainties of the DL models were assessed. Main results. For the full test set, with respect to the MC reference, the predicted prostate D 90 metric had mean differences of −0.64% and 0.08% for the UNet and ResUNet TSE models, respectively. In voxel-by-voxel comparisons, the average global dose difference ratio in the [−1%, 1%] range included 91.0% and 93.0% of voxels for the UNet and the ResUNet TSE, respectively. One forward pass or prediction took 4 ms for a 3D dose volume of 2.56 M voxels (128 × 160 × 128). The ResUNet TSE model closely encoded the well-known physics of the problem as seen in a set of uncertainty maps. The ResUNet TSE rectum D 2cc had the largest uncertainty metric of 0.0042. Significance. The proposed DL models serve as rapid dose predictors that consider the patient anatomy and interseed attenuation effects. The derived uncertainty is interpretable, highlighting areas where DL models may struggle to provide accurate estimations. The uncertainty analysis offers a comprehensive evaluation tool for dose predictor model assessment.
- 20 November 2023
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 68 (23) , 235001
A Monte Carlo dose recalculation pipeline for durable datasets: an I-125 LDR prostate brachytherapy use case
Monte Carlo (MC) dose datasets are valuable for large-scale dosimetric studies. This work aims to build and validate a DICOM-compliant automated MC dose recalculation pipeline with an application to the production of I-125 low dose-rate prostate brachytherapy MC datasets. Built as a self-contained application, the recalculation pipeline ingested clinical DICOM-RT studies, reproduced the treatment into the Monte Carlo simulation, and outputted a traceable and durable dose distribution in the DICOM dose format. MC simulations with TG43-equivalent conditions using both TOPAS and egs_brachy MC codes were compared to TG43 calculations to validate the pipeline. The consistency of the pipeline when generating TG186 simulations was measured by comparing simulations made with both MC codes. Finally, egs_brachy simulations were run on a 240-patient cohort to simulate a large-scale application of the pipeline. Compared to line source TG43 calculations, simulations with both MC codes had more than 90% of voxels with a global difference under ±1%. Differences of 2.1% and less were seen in dosimetric indices when comparing TG186 simulations from both MC codes. The large-scale comparison of egs_brachy simulations with treatment planning system dose calculation seen the same dose overestimation of TG43 calculations showed in previous studies. The MC dose recalculation pipeline built and validated against TG43 calculations in this work efficiently produced durable MC dose datasets. Since the dataset could reproduce previous dosimetric studies within 15 h at a rate of 20 cases per 25 min, the pipeline is a promising tool for future large-scale dosimetric studies.
- 20 November 2023
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 68 (23) , 235001
A Monte Carlo dose recalculation pipeline for durable datasets: an I-125 LDR prostate brachytherapy use case
Monte Carlo (MC) dose datasets are valuable for large-scale dosimetric studies. This work aims to build and validate a DICOM-compliant automated MC dose recalculation pipeline with an application to the production of I-125 low dose-rate prostate brachytherapy MC datasets. Built as a self-contained application, the recalculation pipeline ingested clinical DICOM-RT studies, reproduced the treatment into the Monte Carlo simulation, and outputted a traceable and durable dose distribution in the DICOM dose format. MC simulations with TG43-equivalent conditions using both TOPAS and egs_brachy MC codes were compared to TG43 calculations to validate the pipeline. The consistency of the pipeline when generating TG186 simulations was measured by comparing simulations made with both MC codes. Finally, egs_brachy simulations were run on a 240-patient cohort to simulate a large-scale application of the pipeline. Compared to line source TG43 calculations, simulations with both MC codes had more than 90% of voxels with a global difference under ±1%. Differences of 2.1% and less were seen in dosimetric indices when comparing TG186 simulations from both MC codes. The large-scale comparison of egs_brachy simulations with treatment planning system dose calculation seen the same dose overestimation of TG43 calculations showed in previous studies. The MC dose recalculation pipeline built and validated against TG43 calculations in this work efficiently produced durable MC dose datasets. Since the dataset could reproduce previous dosimetric studies within 15 h at a rate of 20 cases per 25 min, the pipeline is a promising tool for future large-scale dosimetric studies.
- 23 May 2023
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 68 (11) , 114001
Fast D M,M calculation in LDR brachytherapy using deep learning methods
Objective. The Monte Carlo (MC) method provides a complete solution to the tissue heterogeneity effects in low-energy low-dose rate (LDR) brachytherapy. However, long computation times limit the clinical implementation of MC-based treatment planning solutions. This work aims to apply deep learning (DL) methods, specifically a model trained with MC simulations, to predict accurate dose to medium in medium (D M,M) distributions in LDR prostate brachytherapy. Approach. To train the DL model, 2369 single-seed configurations, corresponding to 44 prostate patient plans, were used. These patients underwent LDR brachytherapy treatments in which 125I SelectSeed sources were implanted. For each seed configuration, the patient geometry, the MC dose volume and the single-seed plan volume were used to train a 3D Unet convolutional neural network. Previous knowledge was included in the network as an r 2 kernel related to the first-order dose dependency in brachytherapy. MC and DL dose distributions were compared through the dose maps, isodose lines, and dose-volume histograms. Features enclosed in the model were visualized. Main results. Model features started from the symmetrical kernel and finalized with an anisotropic representation that considered the patient organs and their interfaces, the source position, and the low- and high-dose regions. For a full prostate patient, small differences were seen below the 20% isodose line. When comparing DL-based and MC-based calculations, the predicted CTV D 90 metric had an average difference of −0.1%. Average differences for OARs were −1.3%, 0.07%, and 4.9% for the rectum D 2cc, the bladder D 2cc, and the urethra D 0.1cc. The model took 1.8 ms to predict a complete 3D D M,M volume (1.18 M voxels). Significance. The proposed DL model stands for a simple and fast engine which includes prior physics knowledge of the problem. Such an engine considers the anisotropy of a brachytherapy source and the patient tissue composition.
- 23 May 2023
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 68 (11) , 114001
Fast D M,M calculation in LDR brachytherapy using deep learning methods
Objective. The Monte Carlo (MC) method provides a complete solution to the tissue heterogeneity effects in low-energy low-dose rate (LDR) brachytherapy. However, long computation times limit the clinical implementation of MC-based treatment planning solutions. This work aims to apply deep learning (DL) methods, specifically a model trained with MC simulations, to predict accurate dose to medium in medium (D M,M) distributions in LDR prostate brachytherapy. Approach. To train the DL model, 2369 single-seed configurations, corresponding to 44 prostate patient plans, were used. These patients underwent LDR brachytherapy treatments in which 125I SelectSeed sources were implanted. For each seed configuration, the patient geometry, the MC dose volume and the single-seed plan volume were used to train a 3D Unet convolutional neural network. Previous knowledge was included in the network as an r 2 kernel related to the first-order dose dependency in brachytherapy. MC and DL dose distributions were compared through the dose maps, isodose lines, and dose-volume histograms. Features enclosed in the model were visualized. Main results. Model features started from the symmetrical kernel and finalized with an anisotropic representation that considered the patient organs and their interfaces, the source position, and the low- and high-dose regions. For a full prostate patient, small differences were seen below the 20% isodose line. When comparing DL-based and MC-based calculations, the predicted CTV D 90 metric had an average difference of −0.1%. Average differences for OARs were −1.3%, 0.07%, and 4.9% for the rectum D 2cc, the bladder D 2cc, and the urethra D 0.1cc. The model took 1.8 ms to predict a complete 3D D M,M volume (1.18 M voxels). Significance. The proposed DL model stands for a simple and fast engine which includes prior physics knowledge of the problem. Such an engine considers the anisotropy of a brachytherapy source and the patient tissue composition.
- 16 February 2023
- Published by Elsevier in Physica Medica
- Vol. 107, 102516
Validation of the TOPAS Monte Carlo toolkit for LDR brachytherapy simulations
- 16 February 2023
- Published by Elsevier in Physica Medica
- Vol. 107, 102516
Validation of the TOPAS Monte Carlo toolkit for LDR brachytherapy simulations
- 1 July 2021
- Published by Elsevier in Brachytherapy
- Vol. 20 (4) , 911-921
Validation of the TOPAS Monte Carlo toolkit for HDR brachytherapy simulations
- 1 July 2021
- Published by Elsevier in Brachytherapy
- Vol. 20 (4) , 911-921
Validation of the TOPAS Monte Carlo toolkit for HDR brachytherapy simulations