Do we need patient-specific QA for adaptively generated plans? Retrospective evaluation of delivered online adaptive treatment plans on Varian Ethos.

BACKGROUND: The clinical introduction of dedicated treatment units for online adaptive radiation therapy (OART) has led to widespread adoption of daily adaptive radiotherapy. OART allows for rapid generation of treatment plans using daily patient anatomy, potentially leading to reduction of treatment margins and increased normal tissue sparing. However, the OART workflow does not allow for measurement of patient-specific quality assurance (PSQA) during treatment delivery sessions and instead relies on secondary dose calculations for verification of adapted plans. It remains unknown if independent dose verification is a sufficient surrogate for PSQA measurements. PURPOSE: To evaluate the plan quality of previously treated adaptive plans through multiple standard PSQA measurements. METHODS: This IRB-approved retrospective study included sixteen patients previously treated with OART at our institution. PSQA measurements were performed for each patient's scheduled and adaptive plans: five adaptive plans were randomly selected to perform ion chamber measurements and two adaptive plans were randomly selected for ArcCHECK measurements. The same ArcCHECK 3D dose distribution was also sent to Mobius3D to evaluate the second-check dosimetry system. RESULTS: All (n = 96) ion chamber measurements agreed with the planned dose within 3% with a mean of 1.4% (± 0.7%). All (n = 48) plans passed ArcCHECK measurements using a 95% gamma passing threshold and 3%/2 mm criteria with a mean of 99.1% (± 0.7%). All (n = 48) plans passed Mobius3D second-check performed with 95% gamma passing threshold and 5%/3 mm criteria with a mean of 99.0% (± 0.2%). CONCLUSION: Plan measurement for PSQA may not be necessary for every online-adaptive treatment verification. We recommend the establishment of a periodic PSQA check to better understand trends in passing rates for delivered adaptive treatments.

Cardiac motion and its dosimetric impact during radioablation for refractory ventricular tachycardia.

INTRODUCTION: Cardiac radioablation (CR) is a noninvasive treatment option for patients with refractory ventricular tachycardia (VT) during which high doses of radiation, typically 25 Gy, are delivered to myocardial scar. In this study, we investigate motion from cardiac cycle and evaluate the dosimetric impact in a cohort of patients treated with CR. METHODS: This retrospective study included eight patients treated at our institution who had respiratory-correlated and ECG-gated 4DCT scans acquired within 2 weeks of CR. Deformable image registration was applied between maximum systole (SYS) and diastole (DIAS) CTs to assess cardiac motion. The average respiratory-correlated CT (AVGresp ) was deformably registered to the average cardiac (AVGcardiac ), SYS, and DIAS CTs, and contours were propagated using the deformation vector fields (DVFs). Finally, the original treatment plan was recalculated on the deformed AVGresp CT for dosimetric assessment. RESULTS: Motion magnitudes were measured as the mean (SD) value over the DVFs within each structure. Displacement during the cardiac cycle for all chambers was 1.4 (0.9) mm medially/laterally (ML), 1.6 (1.0) mm anteriorly/posteriorly (AP), and 3.0 (2.8) mm superiorly/inferiorly (SI). Displacement for the 12 distinct clinical target volumes (CTVs) was 1.7 (1.5) mm ML, 2.4 (1.1) mm AP, and 2.1 (1.5) SI. Displacements between the AVGresp and AVGcardiac scans were 4.2 (2.0) mm SI and 5.8 (1.4) mm total. Dose recalculations showed that cardiac motion may impact dosimetry, with dose to 95% of the CTV dropping from 27.0 (1.3) Gy on the AVGresp to 20.5 (7.1) Gy as estimated on the AVGcardiac . CONCLUSIONS: Cardiac CTV motion in this patient cohort is on average below 3 mm, location-dependent, and when not accounted for in treatment planning may impact target coverage. Further study is needed to assess the impact of cardiac motion on clinical outcomes.

Evaluation and correlation of patient movement during online adaptive radiotherapy with CBCT and a surface imaging system.

PURPOSE: With the clinical implementation of kV-CBCT-based daily online-adaptive radiotherapy, the ability to monitor, quantify, and correct patient movement during adaptive sessions is paramount. With sessions lasting between 20-45 min, the ability to detect and correct for small movements without restarting the entire session is critical to the adaptive workflow and dosimetric outcome. The purpose of this study was to quantify and evaluate the correlation of observed patient movement with machine logs and a surface imaging (SI) system during adaptive radiation therapy. METHODS: Treatment machine logs and SGRT registration data log files for 1972 individual sessions were exported and analyzed. For each session, the calculated shifts from a pre-delivery position verification CBCT were extracted from the machine logs and compared to the SGRT registration data log files captured during motion monitoring. The SGRT calculated shifts were compared to the reported shifts of the machine logs for comparison for all patients and eight disease site categories. RESULTS: The average (±STD) net displacement of the SGRT shifts were 2.6 ± 3.4 mm, 2.6 ± 3.5 mm, and 3.0 ± 3.2 in the lateral, longitudinal, and vertical directions, respectively. For the treatment machine logs, the average net displacements in the lateral, longitudinal, and vertical directions were 2.7 ± 3.7 mm, 2.6 ± 3.7 mm, and 3.2 ± 3.6 mm. The average difference (Machine-SGRT) was -0.1 ± 1.8 mm, 0.2 ± 2.1 mm, and -0.5 ± 2.5 mm for the lateral, longitudinal, and vertical directions. On average, a movement of 5.8 ± 5.6 mm and 5.3 ± 4.9 mm was calculated prior to delivery for the CBCT and SGRT systems, respectively. The Pearson correlation coefficient between CBCT and SGRT shifts was r = 0.88. The mean and median difference between the treatment machine logs and SGRT log files was less than 1 mm for all sites. CONCLUSION: Surface imaging should be used to monitor and quantify patient movement during adaptive radiotherapy.

Automated evaluation for rapid implementation of knowledge-based radiotherapy planning models.

PURPOSE: Knowledge-based planning (KBP) offers the ability to predict dose-volume metrics based on information extracted from previous plans, reducing plan variability and improving plan quality. As clinical integration of KBP is increasing there is a growing need for quantitative evaluation of KBP models. A .NET-based application, RapidCompare, was created for automated plan creation and analysis of Varian RapidPlan models. METHODS: RapidCompare was designed to read calculation parameters and a list of reference plans. The tool copies the reference plan field geometry and structure set, applies the RapidPlan model, optimizes the KBP plan, and generates data for quantitative evaluation of dose-volume metrics. A cohort of 85 patients, divided into training (50), testing (10), and validation (25) groups, was used to demonstrate the utility of RapidCompare. After training and tuning, the KBP model was paired with three different optimization templates to compare various planning strategies in the validation cohort. All templates used the same set of constraints for the planning target volume (PTV). For organs-at-risk, the optimization template provided constraints using the whole dose-volume histogram (DVH), fixed-dose/volume points, or generalized equivalent uniform dose (gEUD). The resulting plans from each optimization approach were compared using DVH metrics. RESULTS: RapidCompare allowed for the automated generation of 75 total plans for comparison with limited manual intervention. In comparing optimization techniques, the Dose/Volume and Lines optimization templates generated plans with similar DVH metrics, with a slight preference for the Lines technique with reductions in heart V30Gy and spinal cord max dose. The gEUD model produced high target heterogeneity. CONCLUSION: Automated evaluation allowed for the exploration of multiple optimization templates in a larger validation cohort than would have been feasible using a manual approach. A final KBP model using line optimization objectives produced the highest quality plans without human intervention.

A component method to delineate surgical spine implants for proton Monte Carlo dose calculation.

PURPOSE: Metallic implants have been correlated to local control failure for spinal sarcoma and chordoma patients due to the uncertainty of implant delineation from computed tomography (CT). Such uncertainty can compromise the proton Monte Carlo dose calculation (MCDC) accuracy. A component method is proposed to determine the dimension and volume of the implants from CT images. METHODS: The proposed component method leverages the knowledge of surgical implants from medical supply vendors to predefine accurate contours for each implant component, including tulips, screw bodies, lockers, and rods. A retrospective patient study was conducted to demonstrate the feasibility of the method. The reference implant materials and samples were collected from patient medical records and vendors, Medtronic and NuVasive. Additional CT images with extensive features, such as extended Hounsfield units and various reconstruction diameters, were used to quantify the uncertainty of implant contours. RESULTS: For in vivo patient implant estimation, the reference and the component method differences were 0.35, 0.17, and 0.04 cm3 for tulips, screw bodies, and rods, respectively. The discrepancies by a conventional threshold method were 5.46, 0.76, and 0.05 cm3 , respectively. The mischaracterization of implant materials and dimensions can underdose the clinical target volume coverage by 20 cm3 for a patient with eight lumbar implants. The tulip dominates the dosimetry uncertainty as it can be made from titanium or cobalt-chromium alloys by different vendors. CONCLUSIONS: A component method was developed and demonstrated using phantom and patient studies with implants. The proposed method provides more accurate implant characterization for proton MCDC and can potentially enhance the treatment quality for proton therapy. The current proof-of-concept study is limited to the implant characterization for lumbar spine. Future investigations could be extended to cervical spine and dental implants for head-and-neck patients where tight margins are required to spare organs at risk.

A roadmap for implementation of kV-CBCT online adaptive radiation therapy and initial first year experiences.

PURPOSE: Online Adaptive Radiation Therapy (oART) follows a different treatment paradigm than conventional radiotherapy, and because of this, the resources, implementation, and workflows needed are unique. The purpose of this report is to outline our institution's experience establishing, organizing, and implementing an oART program using the Ethos therapy system. METHODS: We include resources used, operational models utilized, program creation timelines, and our institutional experiences with the implementation and operation of an oART program. Additionally, we provide a detailed summary of our first year's clinical experience where we delivered over 1000 daily adaptive fractions. For all treatments, the different stages of online adaption, primary patient set-up, initial kV-CBCT acquisition, contouring review and edit of influencer structures, target review and edits, plan evaluation and selection, Mobius3D 2nd check and adaptive QA, 2nd kV-CBCT for positional verification, treatment delivery, and patient leaving the room, were analyzed. RESULTS: We retrospectively analyzed data from 97 patients treated from August 2021-August 2022. One thousand six hundred seventy seven individual fractions were treated and analyzed, 632(38%) were non-adaptive and 1045(62%) were adaptive. Seventy four of the 97 patients (76%) were treated with standard fractionation and 23 (24%) received stereotactic treatments. For the adaptive treatments, the generated adaptive plan was selected in 92% of treatments. On average(±std), adaptive sessions took 34.52 ± 11.42 min from start to finish. The entire adaptive process (from start of contour generation to verification CBCT), performed by the physicist (and physician on select days), was 19.84 ± 8.21 min. CONCLUSION: We present our institution's experience commissioning an oART program using the Ethos therapy system. It took us 12 months from project inception to the treatment of our first patient and 12 months to treat 1000 adaptive fractions. Retrospective analysis of delivered fractions showed that the average overall treatment time was approximately 35 min and the average time for the adaptive component of treatment was approximately 20 min.

Nuclear halo measurements for accurate prediction of field size factor in a Varian ProBeam proton PBS system.

PURPOSE: For pencil-beam scanning proton therapy systems, in-air non-Gaussian halo can significantly impact output at small field sizes and low energies. Since the low-intensity tail of spot profile (halo) is not necessarily modeled in treatment planning systems (TPSs), this can potentially lead to significant differences in patient dose distribution. In this work, we report such impact for a Varian ProBeam system. METHODS: We use a pair magnification technique to measure two-dimensional (2D) spot profiles of protons from 70 to 242 MeV at the treatment isocenter and 30 cm upstream of the isocenter. Measurements are made with both Gafchromic film and a scintillator detector coupled to a CCD camera (IBA Lynx). Spot profiles are measured down to 0.01% of their maximum intensity. Field size factors (FSFs) are compared among calculation using measured 2D profiles, calculation using a clinical treatment planning algorithm (Raystation 8A clinical Monte Carlo), and a CC04 small-volume ion chamber. FSFs were measured for square fields of proton energies ranging from 70 to 242 MeV. RESULTS: All film and Lynx measurements agree within 1 mm for full width at half maximum beam intensity. The measured radial spot profiles disagree with simple Gaussian approximations, which are used for modeling in the TPS. FSF measurements show the magnitude of disagreements between beam output in reality and in the TPS without modeling halo. We found that the clinical TPS overestimated output by as much as 6% for small field sizes of 2 cm at the lowest energy of 70 MeV while the film and Lynx measurements agreed within 4% and 1%, respectively, for this FSF. CONCLUSIONS: If the in-air halo for low-energy proton beams is not fully modeled by the TPS, this could potentially lead to under-dosing small, shallow treatment volumes in PBS treatment plans.

Role of phase separation on the biological performance of 45S5 Bioglass®.

We analyzed the biological performance of spinodally and droplet-type phase-separated 45S5 Bioglass® generated by quenching the melt from different equilibrium temperatures. MC3T3-E1 pre-osteoblast cells attached more efficiently to 45S5 Bioglass® with spinodal than to the one with droplet morphology, providing the first demonstration of the role of micro-/nano-scale on the bioactivity of Bioglass®. Upon exposure to biological solutions, phosphate buffered saline (PBS) and cell culture medium (α-MEM), a layer of hydroxyapatite (HA) formed on both glass morphologies. Although both Bioglass® varieties were incubated under identical conditions, and physico-chemical characteristics of the HA layers were similar, the adsorption magnitude of a model protein, bovine serum albumin (BSA, an abundant blood serum component) and its ß-sheet/ß-turn ratio and α-helix content were significantly higher on spinodal than droplet type Bioglass®. These results indicate that: (i) a protein layer quickly adsorbs on the surface of 45S5 Bioglass® varieties (with or without HA layer), (ii) the amount and the conformation of adsorbed proteins are guided by the glass micro-/nano-structure, and (iii) cell attachment and proliferation are influenced by the concentration and the conformation of attached proteins with a significantly better cell adhesion to spinodal type 45S5 Bioglass® substrate. Taken together, our results indicate that the biological performance of 45S5 Bioglass® can be improved further with a relatively simple, inexpensive fabrication procedure that provides a superior glass micro-/nano-structure. A simple modification to the fabrication procedure of classic 45S5 Bioglass® generates spinodal (A(a)) and droplet (A(b)) varieties and has a significant impact on protein adsorption (B) and cell adhesion (C).

Imaging and characterization of optical emission fromex vivotissue during conventional and UHDR PBS proton therapy.

Objective. Imaging of optical photons emitted from tissue during radiotherapy is a promising technique for real-time visualization of treatment delivery, offering applications in dose verification, treatment monitoring, and retrospective treatment plan comparison. This research aims to explore the feasibility of intensified imaging of tissue luminescence during proton therapy (PT), under both conventional and ultra-high dose rate (UHDR) conditions.Approach. Conventional and UHDR pencil beam scanning (PBS) PT irradiation of freshex vivoporcine tissue and tissue-mimicking plastic phantom was imaged using intensified complementary metal-oxide-semiconductor(CMOS) cameras. The optical emission from tissue was characterized during conventional irradiation using both blue and red-sensitive intensifiers to ensure adequate spectral coverage. Spectral characterization was performed using bandpass filters between the lens and sensor. Imaging of conventional proton fields (240 MeV, 10 nA) was performed at 100 Hz frame rate, while UHDR PBS proton delivery (250 MeV, 99 nA) was recorded at 1 kHz frame rate. Dependence of optical emission yield on proton energy was studied using an optical tissue-mimicking plastic phantom and a range shifter. Finally, we demonstrated fast beam tracking capability of fast camera towardsin vivomonitoring of FLASH PT.Main results. Under conventional treatment dose rates optical emission was imaged with single spot resolution. Spot profiles were found to agree with the treatment planning system calculation within >90% for all spectral bands and spot intensity was found to vary with spectral filtration. The resultant polychromatic emission presented a maximum intensity at 650 nm and decreasing signal at lower wavelengths, which is consistent with expected attenuation patterns of high fat and muscle tissue. For UHDR beam imaging, optical yield increased with higher proton energy. Imaging at 1 kHz allowed continuous monitoring of delivery during porcine tissue irradiation, with clear identification of individual dwell positions. The number of dwell positions matched the treatment plan in total and per row showing adequate temporal capability of iCMOS imaging.Significance. For the first time, this study characterizes optical emission from tissue during PT and demonstrates our capability of fast optical tracking of pencil proton beam on the tissue anatomy in both conventional and UHDR setting. Similar to the Cherenkov imaging in radiotherapy, this imaging modality could enable a seamless, independent validation of PT treatments.

Safety and Efficiency Analysis of Operational Decision-Making During Cone Beam Computed Tomography-Based Online Adaptive Radiation Therapy.

PURPOSE: Cone beam computed tomography (CBCT)-based online adaptive radiation therapy (ART) is especially beneficial for patients with large interfractional anatomic changes. However, treatment planning and review decisions need to be made at the treatment console in real-time and may be delegated to clinical staff whose conventional scope of practice does not include making such decisions. Therefore, implementation can create new safety risks and inefficiencies. The objective of this work is to systematically analyze the safety and efficiency implications of human decision-making during the treatment session for CBCT-based online ART. METHODS AND MATERIALS: The analysis was performed by applying the Systems-Theoretical Process Analysis technique and its extension for human decision-making. Four centers of different CBCT-based online ART practice models comprised the analysis team. RESULTS: The general radiation therapy control structure was refined to model the interactions between routine treatment delivery staff and in-person or remote support staff. The treatment delivery staff perform 6 key control actions. Eighteen undesirable states of those control actions were identified as affecting safety and/or efficiency. In turn, 97 hazardous clinical scenarios were identified, with the control action "prepare and position patient" having the least number of scenarios and "delineate/edit influencer and target structures" having the most. Five of these are specific to either in-person or remote support during the treatment session, and 12 arise from staff support in general. CONCLUSIONS: An optimally safe and efficient online ART program should require little to no support staff at the treatment console to reduce staff coordination. Uptraining of the staff already at the treatment console is needed to achieve this goal. Beyond the essential knowledge and skills such as contour editing and the selection of an optimal plan, uptraining should also target the specific cognitive biases identified in this work and the cognitive strategies to overcome these biases. Additionally, technological and organizational changes are necessary.

Unlocking the adaptive advantage: correlation and machine learning classification to identify optimal online adaptive stereotactic partial breast candidates.

Objective.Online adaptive radiotherapy (OART) is a promising technique for delivering stereotactic accelerated partial breast irradiation (APBI), as lumpectomy cavities vary in location and size between simulation and treatment. However, OART is resource-intensive, increasing planning and treatment times and decreasing machine throughput compared to the standard of care (SOC). Thus, it is pertinent to identify high-yield OART candidates to best allocate resources.Approach.Reference plans (plans based on simulation anatomy), SOC plans (reference plans recalculated onto daily anatomy), and daily adaptive plans were analyzed for 31 sequential APBI targets, resulting in the analysis of 333 treatment plans. Spearman correlations between 22 reference plan metrics and 10 adaptive benefits, defined as the difference between mean SOC and delivered metrics, were analyzed to select a univariate predictor of OART benefit. A multivariate logistic regression model was then trained to stratify high- and low-benefit candidates.Main results.Adaptively delivered plans showed dosimetric benefit as compared to SOC plans for most plan metrics, although the degree of adaptive benefit varied per patient. The univariate model showed high likelihood for dosimetric adaptive benefit when the reference plan ipsilateral breast V15Gy exceeds 23.5%. Recursive feature elimination identified 5 metrics that predict high-dosimetric-benefit adaptive patients. Using leave-one-out cross validation, the univariate and multivariate models classified targets with 74.2% and 83.9% accuracy, resulting in improvement in per-fraction adaptive benefit between targets identified as high- and low-yield for 7/10 and 8/10 plan metrics, respectively.Significance.This retrospective, exploratory study demonstrated that dosimetric benefit can be predicted using only ipsilateral breast V15Gy on the reference treatment plan, allowing for a simple, interpretable model. Using multivariate logistic regression for adaptive benefit prediction led to increased accuracy at the cost of a more complicated model. This work presents a methodology for clinics wishing to triage OART resource allocation.

Quantitative, real-time scintillation imaging for experimental comparison of different dose and dose rate estimations in UHDR proton pencil beams.

BACKGROUND: Ultra-high dose rate radiotherapy (UHDR-RT) has demonstrated normal tissue sparing capabilities, termed the FLASH effect; however, available dosimetry tools make it challenging to characterize the UHDR beams with sufficiently high concurrent spatial and temporal resolution. Novel dosimeters are needed for safe clinical implementation and improved understanding of the effect of UHDR-RT. PURPOSE: Ultra-fast scintillation imaging has been shown to provide a unique tool for spatio-temporal dosimetry of conventional cyclotron pencil beam scanning (PBS) deliveries, indicating the potential use for characterization of UHDR PBS proton beams. The goal of this work is to introduce this novel concept and demonstrate its capabilities in recording high-resolution dose rate maps at FLASH-capable proton beam currents, as compared to log-based dose rate calculation, internally developed UHDR beam simulation, and a fast point detector (EDGE diode). METHODS: The light response of a scintillator sheet located at isocenter and irradiated by PBS proton fields (40-210 nA, 250 MeV) was imaged by an ultra-fast iCMOS camera at 4.5-12 kHz sampling frequency. Camera sensor and image intensifier gain were optimized to maximize the dynamic range; the camera acquisition rate was also varied to evaluate the optimal sampling frequency. Large field delivery enabled flat field acquisition for evaluation of system response homogeneity. Image intensity was calibrated to dose with film and the recorded spatio-temporal data was compared to a PPC05 ion chamber, log-based reconstruction, and EDGE diode. Dose and dose rate linearity studies were performed to evaluate agreement under various beam conditions. Calculation of full-field mean and PBS dose rate maps were calculated to highlight the importance of high resolution, full-field information in UHDR studies. RESULTS: Camera response was linear with dose (R2 = 0.997) and current (R22 = 0.98) in the range from 2-22 Gy and 40-210 nA, respectively, when compared to ion chamber readings. The deviation of total irradiation time calculated with the imaging system from the log file recordings decreased from 0.07% to 0.03% when imaging at 12 kfps versus 4.5 kfps. Planned and delivered spot positions agreed within 0.2 ± $\pm$ 0.1 mm and total irradiation time agreed within 0.2 ± $\pm$ 0.2 ms when compared with the log files, indicating the high concurrent spatial and temporal resolution. For all deliveries, the PBS dose rate measured at the diode location agreed between the imaging and the diode within 3% ± $\pm$ 2% and with the simulation within 5% ± $\pm$ 3% CONCLUSIONS: Full-field mapping of dose and dose rate is imperative for complete understanding of UHDR PBS proton dose delivery. The high linearity and various spatiotemporal metric reporting capabilities confirm the continued use of this camera system for UHDR beam characterization, especially for spatially resolved dose rate information.

Improved Dosimetry and Plan Quality for Accelerated Partial Breast Irradiation Using Online Adaptive Radiation Therapy: A Single Institutional Study.

Purpose: Accelerated partial breast irradiation (APBI) is an attractive treatment modality for eligible patients as it has been shown to result in similar local control and improved cosmetic outcomes compared with whole breast radiation therapy. The use of online adaptive radiation therapy (OART) for APBI is promising as it allows for a reduction of planning target volume margins because breast motion and lumpectomy cavity volume changes are accounted for in daily imaging. Here we present a retrospective, single-institution evaluation on the adequacy of kV-cone beam computed tomography (CBCT) OART for APBI treatments. Methods and Materials: Nineteen patients (21 treatment sites) were treated to 30 Gy in 5 fractions between January of 2022 and May of 2023. Time between simulation and treatment, change in gross tumor (ie, lumpectomy cavity) volume, and differences in dose volume histogram metrics with adaption were analyzed. The Wilcoxon paired, nonparametric test was used to test for dose volume histogram metric differences between the scheduled plans (initial plans recalculated on daily CBCT anatomy) and delivered plans, either the scheduled or adapted plan, which was reoptimized using daily anatomy. Results: Median (interquartile range) time from simulation to first treatment was 26 days (21-32 days). During this same time, median gross tumor volume reduction was 16.0% (7.3%-23.9%) relative to simulation volume. Adaptive treatments took 31.3 minutes (27.4-36.6 minutes) from start of CBCT to treatment session end. At treatment, the adaptive plan was selected for 86% (89/103) of evaluable fractions. In evaluating plan quality, 78% of delivered plans met all target, organs at risk, and conformity metrics evaluated, compared with 34% of scheduled plans. Conclusions: Use of OART for stereotactic linac-based APBI allowed for safe, high-quality treatments in this cohort of 21 treatment courses. Although treatment delivery times were longer than traditional stereotactic body treatments, there were notable improvements in plan quality for APBI using OART.

Leveraging intelligent optimization for automated, cardiac-sparing accelerated partial breast treatment planning.

Background: Accelerated partial breast irradiation (APBI) yields similar rates of recurrence and cosmetic outcomes as compared to whole breast radiation therapy (RT) when patients and treatment techniques are appropriately selected. APBI combined with stereotactic body radiation therapy (SBRT) is a promising technique for precisely delivering high levels of radiation while avoiding uninvolved breast tissue. Here we investigate the feasibility of automatically generating high quality APBI plans in the Ethos adaptive workspace with a specific emphasis on sparing the heart. Methods: Nine patients (10 target volumes) were utilized to iteratively tune an Ethos APBI planning template for automatic plan generation. Twenty patients previously treated on a TrueBeam Edge accelerator were then automatically replanned using this template without manual intervention or reoptimization. The unbiased validation cohort Ethos plans were benchmarked via adherence to planning objectives, a comparison of DVH and quality indices against the clinical Edge plans, and qualitative reviews by two board-certified radiation oncologists. Results: 85% (17/20) of automated validation cohort plans met all planning objectives; three plans did not achieve the contralateral lung V1.5Gy objective, but all other objectives were achieved. Compared to the Eclipse generated plans, the proposed Ethos template generated plans with greater evaluation planning target volume (PTV_Eval) V100% coverage (p = 0.01), significantly decreased heart V1.5Gy (p< 0.001), and increased contralateral breast V5Gy, skin D0.01cc, and RTOG conformity index (p = 0.03, p = 0.03, and p = 0.01, respectively). However, only the reduction in heart dose was significant after correcting for multiple testing. Physicist-selected plans were deemed clinically acceptable without modification for 75% and 90% of plans by physicians A and B, respectively. Physicians A and B scored at least one automatically generated plan as clinically acceptable for 100% and 95% of planning intents, respectively. Conclusions: Standard left- and right-sided planning templates automatically generated APBI plans of comparable quality to manually generated plans treated on a stereotactic linear accelerator, with a significant reduction in heart dose compared to Eclipse generated plans. The methods presented in this work elucidate an approach for generating automated, cardiac-sparing APBI treatment plans for daily adaptive RT with high efficiency.

Benchmarking Automated Machine Learning-Enhanced Planning With Ethos Against Manual and Knowledge-Based Planning for Locally Advanced Lung Cancer.

Purpose: Currently, there is insufficient guidance for standard fractionation lung planning using the Varian Ethos adaptive treatment planning system and its unique intelligent optimization engine. Here, we address this gap in knowledge by developing a methodology to automatically generate high-quality Ethos treatment plans for locally advanced lung cancer. Methods and Materials: Fifty patients previously treated with manually generated Eclipse plans for inoperable stage IIIA-IIIC non-small cell lung cancer were included in this institutional review board-approved retrospective study. Fifteen patient plans were used to iteratively optimize a planning template for the Daily Adaptive vs Non-Adaptive External Beam Radiation Therapy With Concurrent Chemotherapy for Locally Advanced Non-Small Cell Lung Cancer: A Prospective Randomized Trial of an Individualized Approach for Toxicity Reduction (ARTIA-Lung); the remaining 35 patients were automatically replanned without intervention. Ethos plan quality was benchmarked against clinical plans and reoptimized knowledge-based RapidPlan (RP) plans, then judged using standard dose-volume histogram metrics, adherence to clinical trial objectives, and qualitative review. Results: Given equal prescription target coverage, Ethos-generated plans showed improved primary and nodal planning target volume V95% coverage (P < .001) and reduced lung gross tumor volume V5 Gy and esophagus D0.03 cc metrics (P ≤ .003) but increased mean esophagus and brachial plexus D0.03 cc metrics (P < .001) compared with RP plans. Eighty percent, 49%, and 51% of Ethos, clinical, and RP plans, respectively, were "per protocol" or met "variation acceptable" ARTIA-Lung planning metrics. Three radiation oncologists qualitatively scored Ethos plans, and 78% of plans were clinically acceptable to all reviewing physicians, with no plans receiving scores requiring major changes. Conclusions: A standard Ethos template produced lung radiation therapy plans with similar quality to RP plans, elucidating a viable approach for automated plan generation in the Ethos adaptive workspace.

Mutual enhancing learning-based automatic segmentation of CT cardiac substructure.

Objective.Current segmentation practice for thoracic cancer RT considers the whole heart as a single organ despite increased risks of cardiac toxicities from irradiation of specific cardiac substructures. Segmenting up to 15 different cardiac substructures can be a very time-intensive process, especially due to their different volume sizes and anatomical variations amongst different patients. In this work, a new deep learning (DL)-based mutual enhancing strategy is introduced for accurate and automatic segmentation, especially of smaller substructures such as coronary arteries.Approach.Our proposed method consists of three subnetworks: retina U-net, classification module, and segmentation module. Retina U-net is used as a backbone network architecture that aims to learn deep features from the whole heart. Whole heart feature maps from retina U-net are then transferred to four different sets of classification modules to generate classification localization maps of coronary arteries, great vessels, chambers of the heart, and valves of the heart. Each classification module is in sync with its corresponding subsequent segmentation module in a bootstrapping manner, allowing them to share their encoding paths to generate a mutual enhancing strategy. We evaluated our method on three different datasets: institutional CT datasets (55 subjects) 2) publicly available Multi-Modality Whole Heart Segmentation (MM-WHS) challenge datasets (120 subjects), and Automated Cardiac Diagnosis Challenge (ACDC) datasets (100 subjects). For institutional datasets, we performed five-fold cross-validation on training data (45 subjects) and performed inference on separate hold-out data (10 subjects). For each subject, 15 cardiac substructures were manually contoured by a resident physician and evaluated by an attending radiation oncologist. For the MM-WHS dataset, we trained the network on 100 datasets and performed an inference on a separate hold-out dataset with 20 subjects, each with 7 cardiac substructures. For ACDC datasets, we performed five-fold cross-validation on 100 datasets, each with 3 cardiac substructures. We compared the proposed method against four different network architectures: 3D U-net, mask R-CNN, mask scoring R-CNN, and proposed network without classification module. Segmentation accuracies were statistically compared through dice similarity coefficient, Jaccard, 95% Hausdorff distance, mean surface distance, root mean square distance, center of mass distance, and volume difference.Main results.The proposed method generated cardiac substructure segmentations with significantly higher accuracy (P < 0.05) for small substructures, especially for coronary arteries such as left anterior descending artery (CA-LADA) and right coronary artery (CA-RCA) in comparison to four competing methods. For large substructures (i.e. chambers of the heart), our method yielded comparable results to mask scoring R-CNN method, resulting in significantly (P < 0.05) improved segmentation accuracy in comparison to 3D U-net and mask R-CNN.Significance.A new DL-based mutual enhancing strategy was introduced for automatic segmentation of cardiac substructures. Overall results of this work demonstrate the ability of the proposed method to improve segmentation accuracies of smaller substructures such as coronary arteries without largely compromising the segmentation accuracies of larger substructures. Fast and accurate segmentations of up to 15 substructures can possibly be used as a tool to rapidly generate substructure segmentations followed by physicians' reviews to improve clinical workflow.

Implementation of a Knowledge-Based Treatment Planning Model for Cardiac-Sparing Lung Radiation Therapy.

PURPOSE: High radiation doses to the heart have been correlated with poor overall survival in patients receiving radiation therapy for stage III non-small cell lung cancer (NSCLC). We built a knowledge-based planning (KBP) tool to limit the dose to the heart during creation of volumetric modulated arc therapy (VMAT) treatment plans for patients being treated to 60 Gy in 30 fractions for stage III NSCLC. METHODS AND MATERIALS: A previous study at our institution retrospectively delineated intracardiac volumes and optimized VMAT treatment plans to reduce dose to these substructures and to the whole heart. Two RapidPlan (RP) KBP models were built from this cohort, 1 model using the clinical plans and a separate model using the cardiac-optimized plans. Using target volumes and 6 organs at risk (OARs), models were trained to generate treatment plans in a semiautomated process. The cardiac-sparing KBP model was tested in the same cohort used for training, and both models were tested on an external validation cohort of 30 patients. RESULTS: Both RP models produced clinically acceptable plans in terms of target coverage, dose uniformity, and dose to OARs. Compared with the previously created cardiac-optimized plans, cardiac-sparing RPs showed significant reductions in the mean dose to the esophagus and lungs while performing similarly or better in all evaluated heart dose metrics. When comparing the 2 models, the cardiac-sparing RP showed reduced (P < .05) heart mean and maximum doses as well as volumes receiving 60 Gy, 50 Gy, and 30 Gy. CONCLUSIONS: By using a set of cardiac-optimized treatment plans for training, the proposed KBP model provided a means to reduce the dose to the heart and its substructures without the need to explicitly delineate cardiac substructures. This tool may offer reduced planning time and improved plan quality and might be used to improve patient outcomes.

Automatic delineation of cardiac substructures using a region-based fully convolutional network.

PURPOSE: Radiation dose to specific cardiac substructures, such as the atria and ventricles, has been linked to post-treatment toxicity and has shown to be more predictive of these toxicities than dose to the whole heart. A deep learning-based algorithm for automatic generation of these contours is proposed to aid in either retrospective or prospective dosimetric studies to better understand the relationship between radiation dose and toxicities. METHODS: The proposed method uses a mask-scoring regional convolutional neural network (RCNN) which consists of five major subnetworks: backbone, regional proposal network (RPN), RCNN head, mask head, and mask-scoring head. Multiscale feature maps are learned from computed tomography (CT) via the backbone network. The RPN utilizes these feature maps to detect the location and region-of-interest (ROI) of all substructures, and the final three subnetworks work in series to extract structural information from these ROIs. The network is trained using 55 patient CT datasets, with 22 patients having contrast scans. Threefold cross validation (CV) is used for evaluation on 45 datasets, and a separate cohort of 10 patients are used for holdout evaluation. The proposed method is compared to a 3D UNet. RESULTS: The proposed method produces contours that are qualitatively similar to the ground truth contours. Quantitatively, the proposed method achieved average Dice score coefficients (DSCs) for the whole heart, chambers, great vessels, coronary arteries, the valves of the heart of 0.96, 0.94, 0.93, 0.66, and 0.77 respectively, outperforming the 3D UNet, which achieved DSCs of 0.92, 0.87, 0.88, 0.48, and 0.59 for the corresponding substructure groups. Mean surface distances (MSDs) between substructures segmented by the proposed method and the ground truth were <2 mm except for the left anterior descending coronary artery and the mitral and tricuspid valves, and <5 mm for all substructures. When dividing results into noncontrast and contrast datasets, the model performed statistically significantly better in terms of DSC, MSD, centroid mean distance (CMD), and volume difference for the chambers and whole heart with contrast. Notably, the presence of contrast did not statistically significantly affect coronary artery segmentation DSC or MSD. After network training, all substructures and the whole heart can be segmented on new datasets in less than 5 s. CONCLUSIONS: A deep learning network was trained for automatic delineation of cardiac substructures based on CT alone. The proposed method can be used as a tool to investigate the relationship between cardiac substructure dose and treatment toxicities.

A novel proton counting detector and method for the validation of tissue and implant material maps for Monte Carlo dose calculation.

The presence of artificial implants complicates the delivery of proton therapy due to inaccurate characterization of both the implant and the surrounding tissues. In this work, we describe a method to characterize implant and human tissue mimicking materials in terms of relative stopping power (RSP) using a novel proton counting detector. Each proton is tracked by directly measuring the deposited energy along the proton track using a fast, pixelated spectral detector AdvaPIX-TPX3 (TPX3). We considered three scenarios to characterize the RSPs. First, in-air measurements were made in the presence of metal rods (Al, Ti and CoCr) and bone. Then, measurements of energy perturbations in the presence of metal implants and bone in an anthropomorphic phantom were performed. Finally, sampling of cumulative stopping power (CSP) of the phantom were made at different locations of the anthropomorphic phantom. CSP and RSP information were extracted from energy spectra at each beam path. To quantify the RSP of metal rods we used the shift in the most probable energy (MPE) of CSP from the reference CSP without a rod. Overall, the RSPs were determined as 1.48, 2.06, 3.08, and 5.53 from in-air measurements; 1.44, 1.97, 2.98, and 5.44 from in-phantom measurements, for bone, Al, Ti and CoCr, respectively. Additionally, we sampled CSP for multiple paths of the anthropomorphic phantom ranging from 18.63 to 25.23 cm deriving RSP of soft tissues and bones in agreement within 1.6% of TOPAS simulations. Using minimum error of these multiple CSP, optimal mass densities were derived for soft tissue and bone and they are within 1% of vendor-provided nominal densities. The preliminary data obtained indicates the proposed novel method can be used for the validation of material and density maps, required by proton Monte Carlo Dose calculation, provided by competing multi-energy computed tomography and metal artifact reduction techniques.

Learning-Based Stopping Power Mapping on Dual-Energy CT for Proton Radiation Therapy.

PURPOSE: Dual-energy computed tomography (DECT) has been used to derive relative stopping power (RSP) maps by obtaining the energy dependence of photon interactions. The DECT-derived RSP maps could potentially be compromised by image noise levels and the severity of artifacts when using physics-based mapping techniques. This work presents a noise-robust learning-based method to predict RSP maps from DECT for proton radiation therapy. MATERIALS AND METHODS: The proposed method uses a residual attention cycle-consistent generative adversarial network to bring DECT-to-RSP mapping close to a 1-to-1 mapping by introducing an inverse RSP-to-DECT mapping. To evaluate the proposed method, we retrospectively investigated 20 head-and-neck cancer patients with DECT scans in proton radiation therapy simulation. Ground truth RSP values were assigned by calculation based on chemical compositions and acted as learning targets in the training process for DECT datasets; they were evaluated against results from the proposed method using a leave-one-out cross-validation strategy. RESULTS: The predicted RSP maps showed an average normalized mean square error of 2.83% across the whole body volume and an average mean error less than 3% in all volumes of interest. With additional simulated noise added in DECT datasets, the proposed method still maintained a comparable performance, while the physics-based stoichiometric method suffered degraded inaccuracy from increased noise level. The average differences from ground truth in dose volume histogram metrics for clinical target volumes were less than 0.2 Gy for D95% and Dmax with no statistical significance. Maximum difference in dose volume histogram metrics of organs at risk was around 1 Gy on average. CONCLUSION: These results strongly indicate the high accuracy of RSP maps predicted by our machine-learning-based method and show its potential feasibility for proton treatment planning and dose calculation.