Commissioning of GPU-based multi-criteria optimizer combined with plan navigation tools for high-dose-rate brachytherapy

Abstract

Purpose Recently, our GPU-based multi-criteria optimization (gMCO) algorithm has been integrated in a graphical user interface (gMCO-GUI) that allows real-time plan navigation through a gMCO-generated set of Pareto-optimal plans for high-dose-rate (HDR) brachytherapy. This work reports on the commissioning of the gMCO algorithm into clinical workflow. Material and methods Our MCO workflow was validated against Oncentra Prostate v. 4.2.2 (OcP) and Oncentra Brachy v. 4.6.0 (OcB). 40 HDR prostate brachytherapy patients (20 with OcP and 20 with OcB) were retrospectively re-planned with gMCO algorithm by generating 2,000 Pareto-optimal plans. A single gMCO treatment plan was exported using gMCO-GUI plan navigation tools. The optimized dwell positions and dwell times of gMCO plans were exported via DICOM RTPLAN files to OcP/OcB, where final dosimetry was calculated. TG43 implementation in gMCO was validated against the consensus data of flexisource. Five analytical shapes were used as the ground truth for volume calculations. Dose-volume histogram (DVH) curves generated by gMCO were compared with the ones generated by OcP/OcB. 3D dose distributions (and isodose lines) were validated against OcP/OcB using dice similarity coefficient (DSC), 95% undirected Hausdorff distance (95% HD), and γ analysis. Results Differences between –0.4% and 0.3% were observed between gMCO calculated dose rates and the flexisource consensus data. gMCO volumes were within ±2% agreement in 3/5 volumes (deviations within –2.9% and 0.1%). For 9 key DVH indices, the differences between gMCO and OcP/OcB were within ±1.2%. Regarding the accuracy of key isodose lines, the mean DSC was greater than 0.98, and the mean 95% HD was below 0.4 mm. The fraction of voxels with γ ≤ 1 was greater than 99% for all cases with 1%/1 mm threshold. Conclusions The GPU-based MCO workflow was successfully integrated into the clinical workflow and validated against OcP and OcB.

Development of patient and catheter specific error thresholds for high dose rate prostate brachytherapy

“…Its function enables uploading of a patient treatment plan and the determination of 1000 Pareto optimal plans in under 10 s 34 . Additionally, it is capable of calculating full dose volume histogram (DVH) curves and the key DVH indices for each plan and has been fully validated against Oncentra Prostate for HDR prostate brachytherapy 37 . Although the gMCO is an optimization tool, no optimizations were performed as part of this study.…”

“…Magnitude of catheter shifts 1-6 mm function enables uploading of a patient treatment plan and the determination of 1000 Pareto optimal plans in under 10 s. 34 Additionally, it is capable of calculating full dose volume histogram (DVH) curves and the key DVH indices for each plan and has been fully validated against Oncentra Prostate for HDR prostate brachytherapy. 37 Although the gMCO is an optimization tool, no optimizations were performed as part of this study. Instead, the gMCO's capability to produce large amounts of plans and simultaneously calculate DVH parameters within seconds was utilized to create an additional module within the gMCO framework.…”

Simultaneous catheter and multicriteria optimization for HDR cervical cancer brachytherapy with a complex intracavity/interstitial applicator

“…Based on a previous commissioning study, and for comparison purposes, the DVH curves of all plans (including the CP) are calculated using gMCO DVH module with Oncentra Brachy (OcB) mimic parameters ON with 100 000 random points/structure, 800 bins, and inter-slice interpolation of the contours. 35 All computations are executed on an Intel(R) Core(TM) i9-10920X CPU (128GB and 24 cores @ 3.5 GHz) and a NVIDIA GeForce RTX(TM) 3090 GPU (ASUS TUF OC edition with 24GB of GDDR6X memory and 10496 CUDA cores).…”

“…A dwell step of 5 mm is used and dwell positions with 5 mm margin above the top surface of the ring applicator are deactivated (T&R catheters) to avoid activation of dwell positions inside the applicator and at its surface as done in clinical planning. Based on a previous commissioning study, and for comparison purposes, the DVH curves of all plans (including the CP) are calculated using gMCO DVH module with Oncentra Brachy (OcB) mimic parameters ON with 100 000 random points/structure, 800 bins, and inter‐slice interpolation of the contours 35 . All computations are executed on an Intel(R) Core(TM) i9‐10920X CPU (128GB and 24 cores @ 3.5 GHz) and a NVIDIA GeForce RTX(TM) 3090 GPU (ASUS TUF OC edition with 24GB of GDDR6X memory and 10496 CUDA cores).…”

A Monte Carlo dose recalculation pipeline for durable datasets: an I-125 LDR prostate brachytherapy use case

“…Given the computational burden of simulating the trajectories of billions of particles, Monte Carlo dose calculation has yet to become a standard in the clinic. GPU-based Monte Carlo codes have achieved computing times allowing a 2% dose uncertainty under 10 s (Bonenfant et al 2015, Lemaréchal et al 2015, which is still too slow for clinical treatment plan optimization, often requiring thousands of iterations (Bélanger et al 2022). On the other hand, Monte Carlo dose distributions can bring valuable information outside of treatment planning.…”