VMAT Treatment Research Articles

Evaluation of Plan Quality and Complexity for Automated Intracranial SBRT VMAT Treatment Plans

The purpose of this study was to evaluate the dosimetric quality and plan complexity of intracranial SBRT VMAT plans generated with an automated treatment planning system (TPS) compared to the standard conventional manually optimized VMAT plans. Ten clinical intracranial conventional SBRT VMAT plans manually generated using our standard TPS, (SE), were replanned using an automated VMAT radiosurgery TPS, Elements Cranial (EC). The plan target volume (PTV), OARs, prescription dose, fractionation, and beam arrangement remained constant between the corresponding plans from both systems. The PTV volume varied from 0.861cc to 40.022cc with prescription doses of 25Gy or 30Gy in 5 fractions. For each plan, PTV mean and max dose were recorded, as well as inverse Paddick conformity index (CI), gradient index (GI), and OAR mean and max doses. A Python script was used to calculate plan complexity metrics, including total MU, complexity metric, aperture irregularity metric, and modulation index total. Statistical significance of the comparisons was determined by performing a paired t-Test with α = 0.05. The results of plan quality and complexity metrics for each planning system are presented in Table 1. Dose conformality of the PTV showed no differences between plans generated with either planning system, with no statistical differences in the CI. The GI showed improvement for the automated plans, which appeared to be due to the system allowing higher PTV mean and max doses to achieve a sharper dose gradient from the target. There was a decrease in MU for the automated plans, which also correlated to statistically lower Complexity, Aperture Irregularity, and Modulation Index metrics. Differences in OAR mean and max doses between both systems were statistically insignificant. Clinically acceptable intracranial SBRT VMAT plans can be generated using an Automated TPS. Plan dosimetric quality with respect to PTV coverage and dose conformality, as well as OAR doses, were equivalent to manual plans. The automated plans were found to be less complex than manual plans, which may be related to the automated system allowing larger target dose inhomogeneity. While statistically significant differences were found between both systems, they may not be clinically relevant, and the differences could be related to the different planning techniques between systems.

EPIDEEP: Using a Deep Learning Model to Predict In Vivo Electronic Portal Imaging Device (EPID) Transit Images

To create a deep-learning model to predict in-vivo electronic portal imaging device (EPID) transit images for IMRT treatments. This model was created to predict in-vivo images to identify machine and patient-related errors that occur during beam delivery and are undetectable with current QA approaches. The deep-learning model can make image predictions much faster than Monte Carlo approaches, making image prediction feasible for application in online adaptive radiotherapy. Additionally, the model does not rely on any proprietary information and can be easily utilized by other clinics. Our approach separates the primary and scatter components of in-vivo transit images. The attenuation of primary radiation reaching the EPID panel is modeled analytically, using attenuation measurements from phantoms of known thicknesses. The scatter component is estimated using a convolutional neural network (CNN). The CNN training uses information from the on-treatment cone-beam CTs (CBCTs), and a pretreatment EPID image with no patient in the beam. We acquired 193 IMRT fields/images from 118 patients previously treated on the Varian Halcyon. Treatment sites included the pelvis, abdomen, lungs, and extremities. CBCTs were collected immediately before treatment, to provide an accurate representation of the anatomy. A 3-channel input image was used, consisting of the pretreatment EPID image, a ray tracing projection through the CBCT to the EPID panel, and a projection to isocenter. Model training:validation:test set ratios were 133:20:40 images. The primary and scatter components are added together to give the predicted transit image. Prediction accuracy was assessed by comparing model-predicted and measured in-vivo EPID images with a 3%/3mm and 5%/3mm gamma pass rate. The gamma pass rate for the patients in the training:validation:test was 91.5%:90.4%:92.1% for 3%/3mm and 96.7%:96.6%:97.0% for 5%/3mm. The model can make image predictions in 20 milliseconds. The poor passing rates of some images may be due to CBCT artifacts and patient motion that occurs between the time of CBCT and treatment. This model can predict in-vivo EPID images with an average gamma pass rate greater than 90%. Image predictions from this model can be used to detect in-vivo treatment errors and changes in patient anatomy, providing an additional layer of patient-specific quality assurance. The speed of image predictions is 20 milliseconds, making use feasible for online adaptive treatments, which currently do not utilize patient-specific measurements of the delivered radiation. Upcoming studies will assess the model's ability in detecting clinically relevant errors and changes in patient anatomy that can adversely affect treatment. Future goals include acquiring more data to further improve the model and extending the model to make predictions for VMAT treatments.

4D Monte Carlo dose reconstructions using surface motion measurements

PurposeTo investigate the feasibility of a 4D Monte Carlo based dose reconstruction method to study the dosimetric impact of respiratory motion using surface motion measurements for patients undergoing VMAT treatments for Non-Small Cell Lung Cancer. MethodsThe 4Ddefdosxyznrc/EGSnrc algorithm was used to reconstruct VMAT doses delivered to the patients using machine log files and respiratory traces measured with the RADPOS 4D dosimetry system. The RADPOS sensor was adhered to the patient’s abdomen prior to each treatment fraction and its position was used as a surrogate for tumour motion. Treatment log files were synchronized with the patient respiratory traces. Patient specific respiratory models were generated from deformable registration of the inhale and exhale 4DCT images and the respiratory traces.The reconstructed doses were compared to planned doses calculated with DOSXYZnrc/EGSnrc on the average-intensity and the exhale phase CT images. ResultsRespiratory motion measurements and log files were acquired for 2 patients over 5 treatment fractions each. The motion was predominantly along the anterior/posterior direction (A/P). The average respiratory amplitudes were 8.7 ± 2.7 mm and 10.0 ± 1.2 mm for Patient 1 and 2, respectively. Both patients displayed inter- and intra-fractional variations in the baseline position. Small inter-fractional differences were observed in the reconstructed doses for each patient. Differences between the reconstructed and planned doses were attributed to differences in organ volumes. ConclusionThe 4D reconstruction method was successfully implemented for the two patients studied. Small differences between the planned and reconstructed doses were observed due to the small tumour motion of these patients.

Fully automated volumetric modulated arc therapy planning for locally advanced rectal cancer: feasibility and efficiency

BackgroundVolumetric modulated arc therapy (VMAT) for locally advanced rectal cancer (LARC) has emerged as a promising technique, but the planning process can be time-consuming and dependent on planner expertise. We aimed to develop a fully automated VMAT planning program for LARC and evaluate its feasibility and efficiency.MethodsA total of 26 LARC patients who received VMAT treatment and the computed tomography (CT) scans were included in this study. Clinical target volumes and organs at risk were contoured by radiation oncologists. The automatic planning program, developed within the Raystation treatment planning system, used scripting capabilities and a Python environment to automate the entire planning process. The automated VMAT plan (auto-VMAT) was created by our automated planning program with the 26 CT scans used in the manual VMAT plan (manual-VMAT) and their regions of interests. Dosimetric parameters and time efficiency were compared between the auto-VMAT and the manual-VMAT created by experienced planners. All results were analyzed using the Wilcoxon signed-rank sum test.ResultsThe auto-VMAT achieved comparable coverage of the target volume while demonstrating improved dose conformity and uniformity compared with the manual-VMAT. V30 and V40 in the small bowel were significantly lower in the auto-VMAT compared with those in the manual-VMAT (p < 0.001 and < 0.001, respectively); the mean dose of the bladder was also significantly reduced in the auto-VMAT (p < 0.001). Furthermore, auto-VMAT plans were consistently generated with less variability in quality. In terms of efficiency, the auto-VMAT markedly reduced the time required for planning and expedited plan approval, with 93% of cases approved within one day.ConclusionWe developed a fully automatic feasible VMAT plan creation program for LARC. The auto-VMAT maintained target coverage while providing organs at risk dose reduction. The developed program dramatically reduced the time to approval.

28 Measurement of Neutron Dose for VMAT and SRS Treatment at Energies Between 6 and 10 MV

28 Measurement of Neutron Dose for VMAT and SRS Treatment at Energies Between 6 and 10 MV

Deep-Learning-Based Dose Predictor for Glioblastoma-Assessing the Sensitivity and Robustness for Dose Awareness in Contouring.

External beam radiation therapy requires a sophisticated and laborious planning procedure. To improve the efficiency and quality of this procedure, machine-learning models that predict these dose distributions were introduced. The most recent dose prediction models are based on deep-learning architectures called 3D U-Nets that give good approximations of the dose in 3D almost instantly. Our purpose was to train such a 3D dose prediction model for glioblastoma VMAT treatment and test its robustness and sensitivity for the purpose of quality assurance of automatic contouring. From a cohort of 125 glioblastoma (GBM) patients, VMAT plans were created according to a clinical protocol. The initial model was trained on a cascaded 3D U-Net. A total of 60 cases were used for training, 15 for validation and 20 for testing. The prediction model was tested for sensitivity to dose changes when subject to realistic contour variations. Additionally, the model was tested for robustness by exposing it to a worst-case test set containing out-of-distribution cases. The initially trained prediction model had a dose score of 0.94 Gy and a mean DVH (dose volume histograms) score for all structures of 1.95 Gy. In terms of sensitivity, the model was able to predict the dose changes that occurred due to the contour variations with a mean error of 1.38 Gy. We obtained a 3D VMAT dose prediction model for GBM with limited data, providing good sensitivity to realistic contour variations. We tested and improved the model's robustness by targeted updates to the training set, making it a useful technique for introducing dose awareness in the contouring evaluation and quality assurance process.

Evaluation of a new radiochromic film dosimeter, Gafchomic EBT4, for VMAT, SABR and HDR treatment delivery verification

Objective. To evaluate a new film for radiotherapy dosimetry, Gafchromic EBT4, compared to the current EBT3. To evaluate dose-response and verify test cases in MV external beam and HDR brachytherapy. Approach. Three lots (batches) of EBT4 and three lots of EBT3 films were calibrated at 6 MV over 0–1200 cGy range, using FilmQAPro software. Signal-to-noise of pixel value, reported dose (RD), and factors affecting dosimetry accuracy were evaluated (rotation of the film at scanning, energy response and post-exposure darkening). Both films were exposed to clinical treatment plans (VMAT prostate, SABR lung, single HDR source dwell, and ‘pseudo’ 3-channel HDR cervix brachytherapy). Film-RD was compared to TPS-calculated dose. Main results. EBT4 calibration curves had characteristics more favourable than EBT3 for radiation dosimetry, with improved signal to noise in film-RD of EBT4 compared to EBT3 (increase of average 46% in red and green channels at 500 cGy). Film rotation at scanning and post-exposure darkening was similar for the two films. The energy response of EBT4 is similar to EBT3. For all clinical case studies, EBT4 provided better agreement with the TPS-planned doses than EBT3. VMAT prostate gamma 3%/3 mm passing rate, EBT4 100.0% compared to EBT3 97.9%; SABR lung gamma 2%/2 mm, EBT4 99.6% and EBT3 97.9%; HDR cervix gamma 3%/2 mm, EBT4 97.7% and EBT3 95.0%. Significance. These results show EBT4 is superior to EBT3 for radiotherapy dosimetry validation of TPS plan delivery. Fundamental improvements in noise profile and calibration curve are reported for EBT4. All clinical test cases showed EBT4 provided equivalent or smaller difference in measured dose to TPS calculated dose than EBT3. Baseline data is presented on the achievable accuracy of film dosimetry in radiotherapy using the new Gafchromic EBT4 film.

Experimental validation of a novel method of dose accumulation for the rectum

Background Dose-surface maps (DSMs) are an increasingly popular tool to evaluate spatial dose-outcome relationships for the rectum. Recently, DSM addition has been proposed as an alternative method of dose accumulation from deformable registration-based techniques. In this study, we performed the first experimental investigation of the accuracy at which DSM accumulation can capture the total dose delivered to a rectum’s surface in the presence of inter-fraction motion. Material and methods A custom PVC rectum phantom capable of representing typical rectum inter-fraction motion and filling variations was constructed for this project. The phantom allowed for the placement of EBT3 film sheets on the representative rectum surface to measure rectum surface dose. A multi-fraction prostate VMAT treatment was designed and delivered to the phantom in a water tank for a variety of inter-fraction motion scenarios. DSMs for each fraction were calculated in two ways using CBCT images acquired during delivery and summed to produce accumulated DSMs. Accumulated DSMs were then compared to film measurements using gamma analysis (3%/2 mm criteria). Similarity of isodose clusters between films and DSMs was also investigated. Results Baseline agreement between film measurements and accumulated DSMs for a stationary rectum was 95.6%. Agreement between film and accumulated DSMs in the presence of different types of inter.-fraction motion was ≥92%, and isodose cluster mean distance to agreement was within 1.5 mm for most scenarios. Overall, DSM accumulation performed the best when using DSMs that accounted for changes in rectum path orientation. Conclusion Dose accumulation performed with DSMs was found to accurately replicate total delivered dose to a rectum phantom in the presence of inter-fraction motion.

Study the Quality of Nasopharyngeal plans Using Evaluation Indexes of IMRT and VMAT Treatment planning Techniques

Study the Quality of Nasopharyngeal plans Using Evaluation Indexes of IMRT and VMAT Treatment planning Techniques

IMRT-VMAT in Locally Advanced Rectal Adenocarcinoma: An Observational Multicenter Prospectic Study on Total Neoadiuvant Radiotherapy According Rapido Trial

Background: The phase III multicenter randomized RAPIDO trial (Rectal Cancer And Preoperative Induction therapy followed by Dedicated Operation) has shown the advantege of Total Neoadjuvant treatment (TNT) with short course radiotherapy (RT). However, data on local control have not been confirmed at 5.6 years median follow-up in favor of the TNT arm because patients from the experimental (EXP) group received more often 3D-CRT (p=0.029). The primary end-point of this multicentric observational prospectic study is to assess the real life safety and efficacy of RAPIDO trial protocol using VMAT and IMRT treatment in all enrolled patients. Moreover FDG-PET and MRI images for predictive features compared with cancer phenotype (status of K-ras and EGFR) and study on tumor microenviroment response to short course RT will be also provided. Methods: Locally advanced rectal cancer patients undergoing TNT as defined in RAPIDO trial will be enrolled and prospectively included in this study. MRI and FDG-PET will be performed to staging and restaging the disease. The tumor microenviroment changes assessment with immunophenotype and soluble PDL1 and on EVCs will be done. For sample size calculation, using OR 0.5, 100 patients will be enrolled into two years. Kaplan -model and proportional Cox method will be adopoted for OS, DFS and predictive factors for local control. Conclusions: The results of this study will be useful to assess prospectively the safety and effectiveness of VMAT and IMRT in short course TNT assuming predictive factors for response through radiomics, metabolomics and immunological features. Trial Registration: COMITATO ETICO UNICO REGIONALE ( CEUR ) N.50/2022; San Carlo Hospital, Potenza, Italy.

Virtual pretreatment patient-specific quality assurance of volumetric modulated arc therapy using deep learning.

Automatic patient-specific quality assurance (PSQA) is recently explored using artificial intelligence approaches, and several studies reported the development of machine learning models for predicting the gamma pass rate (GPR) index only. To develop a novel deep learning approach using a generative adversarial network (GAN) to predict the synthetic measured fluence. A novel training method called "dual training," which involves the training of the encoder and decoder separately, was proposed and evaluated for cycle GAN (cycle-GAN) and conditional GAN (c-GAN). A total of 164 VMAT treatment plans, including 344 arcs (training data: 262, validation data: 30, and testing data: 52) from various treatment sites, were selected for prediction model development. For each patient, portal-dose-image-prediction fluence from TPS was used as input, and measured fluence from EPID was used as output/response for model training. Predicted GPR was derived by comparing the TPS fluence with the synthetic measured fluence generated by the DL models using gamma evaluation of criteria 2%/2 mm. The performance of dual training was compared against the traditional single-training approach. In addition, we also developed a separate classification model specifically designed to detect automatically three types of errors (rotational, translational, and MU-scale) in the synthetic EPID-measured fluence. Overall, the dual training improved the prediction accuracy of both cycle-GAN and c-GAN. Predicted GPR results of single training were within 3% for 71.2% and 78.8% of test cases forcycle-GAN and c-GAN, respectively. Moreover, similar results for dual training were 82.7% and 88.5% for cycle-GAN and c-GAN, respectively. The error detection model showed high classification accuracy (>98%) for detecting errors related to rotational and translational errors. However, it struggled to differentiate the fluences with "MU scale error" from "error-free" fluences. We developed a method to automatically generate the synthetic measured fluence and identify errors within them. The proposed dual training improved the PSQA prediction accuracy of both the GAN models, with c-GAN demonstrating superior performance over the cycle-GAN. Our results indicate that the c-GAN with dual training approach combined with error detection model, can accurately generate the synthetic measured fluence for VMAT PSQA and identify the errors. This approach has the potential to pave the way for virtual patient-specific QA of VMAT treatments.

Influence of physical and radiobiological models on evaluation of radiotherapy treatment fractionation schedules for GBM brain tumours

This study aims to evaluate the radiotherapy planning performance considering the clinical results of glioblastoma multiforme (GBM)-treated patients from both physical and radiobiological perspectives. Radiotherapy plans are done for 20 patients with GBM tumors who have received radical VMAT treatment on Eclipse-13.5. Dosimetric and radiobiological evaluation metrics are estimated to compare the two plans in terms of TCP, NTCP, and P+ in the context of the evaluated physical metrics. Both conventional and hypofractionation give better dose homogeneity and conformity to the tumour. In addition, hypofractionation schedule coverage targets cells with radiation doses higher than conventional schemes. However, both schedules are at acceptable and safe levels. Statistically, the calculated metrics showed good significance between conventional and hypofractionated radiotherapy schedules. The results have led to the safe use of hypofractionated radiotherapy that can be extended for different tumours, which helps in breaking down the waiting list of patients in many hospitals, especially in cases of pandemics such as COVID-19.

Four-dimensional dose reconstruction system for lung cancer VMAT treatment: a 4D phantom study

Four-dimensional dose reconstruction system for lung cancer VMAT treatment: a 4D phantom study

PD-0975 Evaluation and validation of density override in VMAT treatment planning for lung SBRT

PD-0975 Evaluation and validation of density override in VMAT treatment planning for lung SBRT

Helical tomotherapy and two types of volumetric modulated arc therapy: dosimetric and clinical comparison for several cancer sites

Radiotherapy accelerators have undergone continuous technological developments. We investigated the differences between Radixact™ and VMAT treatment plans. Sixty patients were included in this study. Dosimetric comparison between the Radixact™ and VMAT plans was performed for six cancer sites: whole-brain, head and neck, lymphoma, lung, prostate, and rectum. The VMAT plans were generated with two Elekta linear accelerators (Synergy®and Versa HD™). The planning target volume (PTV) coverage, organs-at-risk dose constraints, and four dosimetric indexes were considered. The deliverability of the plans was assessed using quality assurance (gamma index evaluation) measurements; clinical judgment was included in the assessment. The mean AAPM TG218 (3%-2mm, global normalization) gamma index valueswere 99.4%, 97.8%, and 96.6% forRadixact™,Versa HD™, andSynergy®, respectively. Radixact™ performed better than Versa HD™ in terms of dosimetric indexes, hippocampi D100%, spinal cord Dmax, rectum V38.4 Gy, bladder V30 Gy,and V40 Gy. Versa HD™saved more of the (lungs-PTV) V5 Gyand (lungs-PTV) Dmean,heartDmean, breasts V4 Gy, and bowel V45 Gy. RegardingSynergy®, the head and neckRadixact™plan saved more of the parotid gland, oral cavity, and supraglottic larynx. From a clinical point of view, for the head and neck, prostate, and rectal sites, theRadixact™ and Versa HD™ plans were similar;Radixact™ plans were preferable for the head and neck and rectum toSynergy®plans. The quality of linac plans has improved, and differences with tomotherapy have decreased. However, tomotherapy continues to be an essential add-on in multi-machine departments.

Clinical implementation of a log file-based machine and patient QA system for IMRT and VMAT treatment plans

PurposeTo determine the error detection sensitivity of a commercial log file-based system (LINACWatch®, LW) for integration into clinical routine and to compare it with a measurement device (OCTAVIUS 4D, Oct4D) for IMRT and VMAT delivery QA. Materials and methods76 VMAT/IMRT plans (H&N, prostate, rectum and breast) preliminarily classified according to their Modulation Complexity Score (MCS) calculated by LW, were considered. Receiver Operating Characteristic (ROC) Curves were used to establish gamma criteria for LW. 12 plans (3 for each site) were intentionally modified in order to introduce delivery errors regarding MLC, jaws, collimator, gantry and MU (for a total set of 168 incorrect plans) and irradiated on Oct4D; the corresponding log files were analysed by LW. Each incorrect plan was compared to the error-free plan using γ-index analysis for MLC, jaws and MU errors investigation and Root-Mean-Square (RMS) values for gantry and collimator errors investigation. ResultsMCS ranges values were: 0.10–0.20 for H&N, 0.21–0.40 for prostate and rectum, 0.41–1.00 for breast. From ROC curves, the Gamma Passing Rate (GPR) thresholds were: 87%, 92%, 99% for H&N, prostate and rectum, and breast, respectively. The 1.5%/1.5 mm/local criteria were adopted for the γ-analysis. LW sensitivity in detecting the introduced errors was higher when compared to Oct4D: 48.5% vs 30.4% respectively. ConclusionsLW can be considered useful complement to phantom-based delivery QA of IMRT/VMAT plans. The MCS tool is effective in detecting over or under modulated plans prior to pre-treatment QA. However, rigorous and routinely machine QCs are recommended.

VMAT partial arc technique decreases dose to organs at risk in whole pelvic radiotherapy for prostate cancer when compared to full arc VMAT and IMRT.

Whole pelvic radiotherapy (WPRT) can sterilize microscopic lymph node metastases in treatment of prostate cancer. WPRT, compared to prostate only radiotherapy (PORT), is associated with increased acute gastrointestinal, and hematological toxicities. To further explore minimizing normal tissue toxicities associated with WPRT in definitive IMRT for prostate cancer, this planning study compared dosimetric differences between static 9-field-IMRT, full arc VMAT, and mixed partial-full arc VMAT techniques. In this retrospective study, 12 prostate cancer patients who met the criteria for WPRT were randomly selected for this study. The initial volume, PTV46, included the prostate, seminal vesicles, and pelvic nodes with margin and was prescribed to 4600cGy. The cone-down volume, PTV78, included the prostate and proximal seminal vesicles with margin to a total dose of 7800cGy. For each CT image set, 3 plans were generated for each of the PTVs: an IMRT plan, a full arc (FA) VMAT plan, and a mixed partial-full arc (PFA) VMAT plan, using 6MV photons energy. According to RTOG protocols none of the plans had a major Conformity Index (CI) violation by any of the 3 planning techniques. PFA plan had the best mean CI index of 1.00 and significantly better than IMRT (p = 0.03) and FA (p = 0.007). For equivalent PTV coverage, the average composite gradient index of the PFA plans was better than the IMRT and the FA plans with values 1.92, 2.03, and 2.01 respectively. The defference was statistically significant between PFA/IMRT and PFA/FA, with p- values of < 0.001. The IMRT plans and the PFA plans provided very similar doses to the rectum, bladder, sigmoid colon, and femoral heads, which were lower than the dose in the FA plans. There was a significant decrease in the mean dose to the rectum from 4524 cGy with the FA to 4182cGy with the PFA and 4091cGy with IMRT (p < 0.001). The percent of rectum receiving 4000cGy was also the highest with FA at 66.1% compared to 49.9% (PFA) and 47.5% (IMRT). There was a significant decrease in the mean dose to the bladder from 3922cGy (FA) to 3551cGy (PFA) and 3612cGy (IMRT) (p < 0.001). The percent of bladder receiving 4000cGy was also the highest with FA at 45.4% compared to 36.6% (PFA) and 37.4% (IMRT). The average mean dose to the sigmoid colon decreased from 4177cGy (FA) to 3893cGy (PFA) and 3819cGy (IMRT). The average mean dose to the femoral heads decreased from 2091cGy (FA) to 2026cGy (PFA) and 1987cGy (IMRT). Considering the improvement in plan quality indices recorded in this study including the dose gradient and the dose to organs at risk, mixed partial-full arc plans may be the preferred VMAT treatment technique over full arc plans for prostate cancer treatments that include nodal volumes.

14P The impact of the deep inspiration breath-hold (DIBH) technique in the heart and other organ-at-risk (OAR) dose sparing in the postoperative radiotherapy to the left early breast cancer

Under the deep inspiration breath hold (DIBH) technique a considerable volume of the heart can be reduced to the high radiation doses. To investigate whether implementing DIBH to the hybrid technique will further reduce the cardiac doses, we compared the dosimetry characteristics of hybrid and VMAT treatment planning under DIBH and free breath.

Clinical Evaluation of an Automated Treatment Planning Framework for Radiation Oncology

<h3>Purpose/Objective(s)</h3> Automation of radiotherapy treatment planning increases consistency, provides a safety net around error prevention, and reduces planning time. The objective of this study was to validate a scalable framework to automate the creation of VMAT and IMRT treatment plans using commercial planning systems. <h3>Materials/Methods</h3> A total of 35 centers across Australia were used for the study and the disease sites consisted of Urinogenital, Lung, Breast, Head and Neck, Neurological, GI and Gynecological. A scalable framework for external beam treatment planning was implemented for two commercial treatment planning systems (TPS), Eclipse (Varian Medical Systems, Palo Alto, CA) and Monaco (Elekta AB, Stockholm, Sweden). The framework received CT images, target volumes, normal tissue structures, and prescription data from dosimetrists. Information was parsed into the system, validated against configuration settings, the desired treatment protocol template selected, optimization parameters adjusted, and dose calculation completed. The framework controlled the TPS through the Application Programming Interface (API). The dosimetry team evaluated each automated plan by comparison of the plan metrics to the prescription and assessment of the dose distribution. The plans were scored in three areas: 1) clinician approved with no intervention, 2) minimal intervention with minor adjustments, and 3) manual replanning. <h3>Results</h3> Since Aug. 2021, 4972 plans (36% of total plans) have been successfully processed across 35 treatment centers with 111 clinicians. 487 plans were rejected by the framework due to error. The system automated plans for GUI (1822), Lung (629), Breast (745), Head and Neck (65), Neurological (96), GI (568) and Gynecological (163). Out of these plans, 23% were clinician approved with no user intervention, 41% required minimal intervention to achieve clinician approval, and 36% required additional optimization structures and/or cycles of optimization to achieve clinician approval. The average time between clinician target volume delineation and plan approval reduced by 25% from 1.6 to 1.2 days. In Feb. 2022, the framework successfully processed 970 plans (41% of total plans), rejecting 126 (5%) due to error. Evaluation between systems showed 27% of Eclipse plans and 43% of Monaco plans were processed successfully. Of these plans, 19% (Eclipse) and 23% (Monaco) required no intervention while 25% (Eclipse) and 46% (Monaco) required minimal intervention. Analysis of dosimetrist productivity showed an increase of 16% since Aug. 2021 while the average time saved per plan was 85 minutes. Analysis of adverse and near- miss dosimetry related events showed a 6% reduction in reported incidents. <h3>Conclusion</h3> The framework for automated planning achieved a 60% clinician approval rate for plans with minimal or no intervention. Automation can result in more consistent planning and timesaving's for expert operators. A 6% reduction in recorded dosimetry related errors was achieved.

Dosimetric Analysis of Radiation Treatment Plans Based on a Deep Learning Auto Contouring Model for Patients with Localized Prostate Cancer

<h3>Purpose/Objective(s)</h3> To evaluate the feasibility and safety of creating radiation treatment plans based on a convolutional neural network auto-contouring (AC) model for patients receiving radiation for localized prostate cancer with a radio-opaque rectal spacer hydrogel. <h3>Materials/Methods</h3> The deep learning model produces ACs for target volumes (prostate, proximal seminal vesicle), OARs (rectum, bladder, penile bulb, femoral heads), and rectal hydrogel spacers. ACs were compared to manual contours (MC) by MDs for clinical use. Individual volumes and overall performance (composite score) were previously evaluated and reported, qualitatively using a 1 (minor discrepancy), 2 (moderate discrepancy), 3 (significant discrepancy), and 4 (rejected, gross error) scale and quantitatively using Dice Similarity Coefficient (DSC) and mean distance to agreement. VMAT treatment plans were created on unedited AC volumes from five composite 1 (C1) cases and five composite 2 (C2) cases and dose volume metrics were computed for the final MCs. Dosimetry comparison was performed to assess the safety of delivering a treatment plan created from unedited ACs. Cases were planned to 50.4 Gy in 28 fractions to the SVs, with a SIB of 70 Gy the prostate. PTV expansions were CTV+5 mm (CTV+4 mm posteriorly). <h3>Results</h3> Of 62 initial cases evaluated, 11 were scored as a C1, 28 were scored as a C2, and 23 received a C3. 10 cases were randomly selected for treatment planning. For prostate (CTV P) in selected C1 and C2 cases, mean score was 1 and 1.8 and mean DSC was 0.919 and 0.876, respectively. Clinical goals for planning and dosimetric data from treatment plans are shown in Table 1. The mean V70 Gy for prostate CTV/PTV was 99.83%/90.05% for C1 cases and 99.46%/89.15% for C2 cases. All clinical goals for OARs were met in both cohorts. <h3>Conclusion</h3> 63% of all cases evaluated received a C1 or C2 score. Our results indicate the treatment plans generated from unedited ACs on these cases respected all clinical goals for OARs. Although C1 plans produced better PTV coverage than C2, almost all plans had sub-optimal PTV coverage, indicating the need for MDs to edit target volumes before treatment planning.