AAPM Spring Clinical Meeting - Abstract

2021 Journal of Applied Clinical Medical Physics  
Purpose: CTDIvol is the current standard measure of CT dosimetry which requires each measurement at a single axial scan. However, modern scanners use helical protocols more than axial protocols, and it is tedious or sometimes problematic to convert helical protocols into equivalent axial protocols. A helical CTDIvol measurement method has been proposed and showed promising results to replace the current standard axial method. However, it was unknown if the method can be utilized for dual energy
more » ... CT helical acquisitions. In this study, the feasibility of the direct helical acquisition CTDIvol measurement method was evaluated with dual energy protocols. Methods: 21 Clinical dual energy protocols from two GE CT scanners were evaluated for the helical CTDIvol method. CTDI phantoms of 16 and 32 cm diameters and a CT ion chamber were used for the measurement. The scan range was prescribed as 10 cm to cover the CT ion chamber. Air kerma readings from the ion chamber were used to calculate CTDIvol. To evaluate the performance of the proposed method, the discrepancy and repeatability indices were assessed along with other statistical analysis including linear regression, Paired Student t-test, and correlation analysis. Results: Paired Student t-test shows the direct helical measurement has no statistically significant difference with the manufacturer reported CTDIvol (p=0.51), exhibiting the absolute discrepancy of -0.12±0.93 (mGy) (mean±standard deviation), discrepancy index of 1.5±3.7 (%) and repeatability index of 98.5±1.5 (%). The linear regression model indicates there is a good linear fitting (y=1.0286x-0.946) between the measured and displayed CTDIvol. No correlation is found between the discrepancy/repeatability indices and pitch, collimation widths, or bowtie filters (p>0.05). Conclusion: The result suggests that the direct helical CTDIvol measurement method is a good alternative for the axial CTDIvol method in evaluating CTDIvol of dual energy CT scanners during quality assurance testing. Purpose: To investigate and compare digital x-ray tomosynthesis (DT) to 2-view radiography (DR) and CT as a candidate modality for detecting tungsten carbide, steel, and aluminum fragments in the orbits for orbital metal screening prior to MRI. Methods: A semi-anthropomorphic head phantom was created by placing a human skull model into a water bath. Shaving fragments of aluminum, steel, and tungsten carbide were created. Varying sized fragments of aluminum (0.1-0.7mg, 0.04-0.26mm³), steel (0.4-0.7mg, 0.05-0.09mm³), and tungsten carbide (1.3-8.1mg, 0.08-0.52mm³) were embedded in table grapes, with one fragment in each grape. The grapes were placed into the skull and images were acquired with our institution's orbit CT protocol, 2-view facial bone DR protocol, and with a facial DT protocol. Resulting images were reviewed by a radiologist to qualitatively evaluate comparative detectability on each modality. For each set of grapes with different fragments embedded, repeatability was investigated by acquiring repeat images after repositioning of the grapes, the skull, and/or no repositioning. Results: For the fragment sizes investigated, tungsten carbide was detectable on all modalities. Steel fragments were detectable on CT, and the largest steel fragment was just detectable on DT but not DR. The largest aluminum fragment was just detectable on CT but not on DT or DR. Repeatability tests indicated consistency in detectability of fragments on CT, while detection of fragments on DT was more position dependent. Conclusion: With respect to detectability of orbital metal fragments, DT performance was found to lie in between that of DR and CT. DT may be an appropriate modality for orbital metal screening, providing a balance between the radiation dose, detection capabilities, and cost of DR and CT. Expansion of this work into a formal detectability study with multiple observers and a wider range of fragment sizes is warranted. Purpose: To investigate and demonstrate potential pitfalls in implementation of CT organ dose calculations in clinical settings. Methods: Our hospital system uses a commercial radiation dose monitoring software to collect clinical radiation dose data from all CT studies. The software acquires and stores radiation dose data from DICOM-compliant radiation dose structured reports (RDSRs) and provides customizable dose reports, high-dose alerts, and many data analysis tools. The software also includes an NCICT Organ Dose calculator, which estimates organ dose and effective dose resulting from patient CT studies, given such details as patient size, CTDIvol and scan region, while matching patient height and weight to a digital phantom in the NCICT phantom library. The body part used for organ dose calculation initially came from the acquisition target region indicated in the RDSR. A simple change was made in the software to instead use the configurable body part associated with the Study Common Name. An audit of the four most common studies was performed, accuracy of scan region was evaluated, and organ doses were recalculated. Results: For each Study Common Name, acquisition target region varied drastically with scanner and workflow. Protocols that spanned multiple body parts had high variability in target regions, where scan regions were only correctly identified in 33% for Body/Trauma and 66% for Abdomen/Pelvis. The configurable body part improved consistency of organ dose calculations. All C-Spine protocols initially had acquisition target regions of neck or head. After adjustment of body part from head to c-spine, the effective dose increased from 3.8 mSv to 6.89 mSv. Conclusion: Organ dose calculations for CT can be improved by properly defining scan region based on Study Common Name body part rather than acquisition target region. Purpose: Remote working developed quickly during the pandemic, and many radiologists had a workstation setup at home to maintain social distancing and to work safely. This study investigates how the setup of the home reading environment can impact low-contrast visibility. Methods: To understand the factors that impact the reading, five basic factors are considered: (1) The room layout, such as window/light fixture and location, (2) the ambient light level, (3) monitor surface cleanness, (4) clutter in the background, and (5) the reader outfit. Ambient light levels were measured by affordable lux meters with in-house calibration. To quantify the impact of these five factors on low-contrast visibility, a letter-based test pattern was developed to assess minimum detectable contrast at different display background luminance levels. The test pattern was displayed on a consumer-grade monitor, and baseline low-contrast visibility was established in an appropriate reading environment. Results: Calibration of the affordable lux meter against a NIST traceable calibrated meter showed a linear relationship with a factor of 0.73 (within 10% variances). Increased ambient light level, external light, and specular reflections, and stains made it difficult to see the low contrast letters in dark and white backgrounds. This indicates that radiologists might miss abnormalities when reading under these conditions. Conclusion: Our results indicate that in-house calibration of an affordable lux meter is feasible, which enables distribution to radiologist reading at home to ensure appropriate ambient light levels. Ambient light not only increases diffuse reflection and degrades the DICOM-14 calibration, but also increases the conspicuity of distracting specular reflections and dirt on display surfaces. The monitor should place in a proper place avoiding reflections from light sources. Also, dark outfits are recommended. A simple step-by-step setup checks minimizing these five factors can help ensure the quality and consistency of remote reading. PO-BPC-Virtual-05 Adapting Mathematical Observer Models for Dermoscopy: A Study V Vasudev 1, 2 *, T Kimpe 1 , (1) BARCO NV, Kortrijk, BE, (2) imec-IPI, Ghent University, BE Purpose: Explore the possible use of mathematical observer models in dermoscopy. Methods: Mathematical Observers (MOs) have been used extensively in a Virtual Clinical Trial context as surrogates for human observers. They have commonly been used for radiology imaging to model lesion detection tasks, for e.g. in mammography and lung cancer screening. Increasingly reliable MOs have been researched that account for limitations of the human visual system (contrast sensitivity function and dynamic adaptation of the eye) as well as complex lesion characteristics and shapes (spiculations).State-of-the-art MOs, commonly used for radiology, have been applied to dermoscopy images and their usefulness and limitations explored. Results: Applying existing MO models, such as channelized hotelling observers, to dermoscopy is not straightforward and does not provide reliable results. Compared to radiology, dermoscopy images are color images so lesions are characterized by changes in color as opposed to intensity (as is the case for most state-of-the-art MO models). Little research has been performed on color MO models. In addition, skin lesions are complex and contain a wide range of features (e.g. globules, white veil, etc). An approach where a MO is trained to detect a single feature is not feasible.Instead, a novel architecture for a layered observer model is proposed, which contains a set of observer models, each targeting a feature/set of features. The individual observer outputs are then combined into an overall observer output.This can be applied to dermoscopy where human observer decisions are typically not based on the presence of individual lesion features, but rather by looking at overall lesion characteristics and presence or absence of groups of features. Conclusion: In this work we outline prior MO model work and explore usefulness and limitations for use in dermoscopy imaging, while exploring adaptations for making MOs better suitable for use in dermoscopy imaging. Purpose: Professionalism is one of the CAMPEP standards required for accreditation of medical physics residency training. Given the lack of literature and resources regarding formalized professionalism training and its definition in medical physics, this work aims to design an effective survey to collect data from program directors (PDs) on the current status of formalized professionalism training. Methods: A systematic approach was implemented during the survey design process. A working group of 5 AAPM-MPRTP/SDAMPP committee members interested in professionalism training was formed. A short pilot survey was sent out to 15 PDs to assess the need (i.e., definition of professionalism, methods for teaching it, skill assessment, and associated challenges). This prompted the working group to design an official survey via a consensus-based approach. Upon completion, the survey was first validated using cognitive pre-testing with an independent reviewer over a qualitative discussion of verbal-probing and think-aloud techniques. Next, quantitative validation was performed by six independent content experts with questions rated on a scale from 1-4 each of increasing relevance and clarity. Metrics known as the reliability agreement (RA), the degree to which interrater ratings agree with each other, and the content validity index (CVI), which measures the collective validity of the expert's rating per question, were calculated from the ratings, each for relevance and clarity. Based on published guidelines, a CVI>0.8 indicates an effective survey question. Results: Pre-survey results indicated wide variability in defining professionalism and the need for formalized training and resources. An official survey of 20 questions divided into 6 sections was developed for validation. In regards to relevance, the average RA=85% and CVI=98%. For clarity, the average RA=80% and CVI=97%. Conclusion: Validation results demonstrated the success of a consensus-based approach for survey question design. Final survey revisions will incorporate validation results and feedback prior to distribution to PDs. Purpose: To quantify the impact of organ deformation on the ablative treatment for pancreatic cancer. Methods: We retrospectively analyzed fourteen patients treated with volumetrically-modulated arc therapy in 2017-2019 for pancreatic cancer. Thirteen patients each had one or more fiducial markers attached to the target and matched by image guidance for treatments. The other one patient was treated with bone match. We generated 72 virtual computed-tomography scans (vCT) by deforming planning CTs (pCT) to cone beam CTs (CBCT). Areas that were matched for treatment were used to guide the deformable registration. Daily dose distributions on the vCT scans were calculated and mapped back to the pCT, where we characterized the daily target coverage by the percentage of gross tumor volume receiving the full dose. The daily dose distributions of each patient were further summed up allowing evaluation of cumulative coverage and other relevant dosimetric quantities, which were compared with the treatment plans. Results: The median daily coverage was 65% (range 12-74%). The daily coverage varied by <10% (standard deviation, SD) for all patients except the bone-matching patient, for whom the SD was 29%. The median cumulative coverage was 58% (range 6-74%), compared to median 79% (range 10-96%) from the original plans. The cumulative target coverage of the bone-matching patient degraded by the largest amount (88% drop). For that patient, the cumulative stomach dose was 62 Gy, which was 5 Gy higher than the planned value and exceeded our limit of 60 Gy. For another patient, the image registration was significantly affected by the CBCT artifacts. Conclusion: Organ deformation significantly affected target coverage for ablative treatment of pancreatic cancer. Daily image guidance with internal marker localization provided some mitigation but is limited by the drawbacks of CBCT imaging. Purpose: To survey deep learning and image segmentation techniques that automatically delineate multiple brain metastases within 3D MRI images. Methods: Six different techniques using deep learning and automatic image segmentation of multiple brain metastases detected in 3D MRI images were found in the literature. The following results were collected from each article: depth of deep learning algorithm, segmentation algorithm, DNN activation function, size of tumors, sensitivity and dice similarity performance scores, and final output produced. Results: Deep learning neural networks, DNN, ranged in depth from 11 layers to 101 layers. The reported sensitivity and false positive rates ranged from 0.77 to 0.98 and the reported dice similarity coefficients ranged from 0.70 to 0.93. Brain metastases smaller than 0.40ml were reported challenging to detect and delineate automatically with most. One solution split the segmentation process into multiple sizes and then applied classification algorithm, En-DeepMedic. Additionally, each study independently built training data from retrospective patient chart review, as there is no benchmark datasets available for this highly specialized radiation therapy field. When results were reported, the range of sizes of the brain metastases were different as they were unique to patients sampled. Conclusion: In this research, data augmentation plays an important role in the training process for treatment site. No one automatic segmentation technique had been adopted for SRS. The solution to this problem continues to evolve with technological advances in computer vision and multiple object detection. The most recent algorithm surveyed, Mask R-CNN, has not been thoroughly researched as a viable solution. Purpose: To compare the treatment plans made by using two different planning systems in which 2x2 cm² and 0.5x0.5 cm² data were entered as the smallest field sizes. Methods: Ten brain metastasis patients with an average PTV volume of 0.17 cm³ were included in the study. The plans of all patients were created by using dynamic conformal arc, 6FFF photon energy, which was shaped by MLC by keeping the 2x2 jaw size constant in the Eclipse treatment planning system (TPS). Dynamic conformal arc treatment plans were created using the same table, collimator and gantry angles and energy in the Iplan TPS. Planning goal was to cover 98 % of the PTV with 98% of the prescribed dose. Acuros (version 13.6) algorithm was used in the Eclipse TPS. Pencil beam (PB) and Monte Carlo (MC) algorithms were used in the Brainlab-iPlan TPS(version 4.4.5). In treatment plans, tissue volume receiving the prescribed dose (V100), PTV receiving the prescribed dose (PTV V100), Conformity Index (CI), Gradient Index (GI), volume of the tissue receiving the prescribed dose (V50), PTVmax, PTVmin, PTVmean dose values, treatment time (MU), brain tissue volume receiving 10Gy (BV10Gy) were compared. The differences were evaluated by comparing Eclipse Acuros with Iplan PB, Acuros (Acu), and MC and PB-MC algorithms. Results: When Acuros and PB and Acu-MC algorithms were compared, no significant differences were found between tissue V100, PTV V100, and CI values. Significant differences were found between V50, GI, PTVmax, PTVmean MU, and BV10Gy doses. p values of the comparisons are presented in Table 1 . Conclusion: Dose distributions varied significantly for field sizes smaller than 2x2cm² created by MLCs. In addition to the appropriate dose calculation algorithms of small volume targets, the use of beam data by small field dosimetry contributes significantly to obtaining more precise and accurate dose calculation results. Purpose: GammaPod is an emerging gamma irradiation device dedicated to breast stereotactic radiotherapy. It contains 25 rotary cobalt-60 sources collimated by 15 or 25-mm collimators to deliver highly-conformal dose for boost, partial-breast irradiation, or neo-adjuvant stereotactic radiosurgery. Annual quality assurance of this system using radiochromic films is the current practice, but it is laborious and time-consuming. This study aims to develop a quality assurance program that utilizes the SRS MapCHECK device (Sun Nuclear Corp, Melbourne, FL). Methods: A custom-designed plastic holder was used to accommodate the SRS MapCHECK embedded in the StereoPHAN™ at the treatment location on the GammaPod™ (Xcision Medical Systems, LLC, Columbia, MD). Single-shot irradiations were used in three preliminary tests: 1). reproducibility of the MapCHECK dose measurement in two consecutive deliveries; 2). dose profile constancy check compared to the baseline measurements; 3). couch motion accuracy by measuring shifts in dose maps with and without couch motions. Results: Within the radiation field, the dose maps from two consecutive measurements were similar with a mean difference of -0.07% (range: -0.21-0.07%) for the 15-mm collimators, and 0.11% (range: -0.13-0.28%) for the 25-mm collimators. Dose profiles along the y and z directions resembled the basal profiles, all with 100% gamma passing rate at 3%/0.5 mm. The spatial shifts quantified from the two dose maps sensitively detected 0.3-mm couch motions, and accurately delineated couch motions at 21.0 mm. Conclusion: This pilot study successfully demonstrated the feasibility and reproducibility of quality assurance of the GammaPod™ system using the SRS MapCHECK. The utility of this device for dose profile constancy check and couch motion tests has also been demonstrated. Ongoing studies are being conducted to investigate its utility in quality assurance of patient-specific plans. Purpose: Deep inspiration breath-hold (DIBH) for left-sided breast/chest-wall irradiation results in significant heart and lung sparing. A bilateral comprehensive irradiation of the chest-wall and locoregional nodes presents a much more challenging case where DIBH is also of benefit. We have developed a robust dual-isocenter DIBH VMAT technique with visual guidance for maximum sparing of the heart and lungs. Methods: A comprehensive chest-wall plan to irradiate the bilateral chest-wall and locoregional nodes was created using a hypofractionated approach to 4256 cGy in 16 fractions. A total of eight partial arcs in two groups of four were optimized simultaneously using automatic feathering to avoid hot/cold spots in the overlap region. The isocenter for each group of partial arcs was placed just posterior to each side of the chest-wall target to accomplish a tangential irradiation. A 1 cm virtual skin flash bolus (HU=-500) was used for optimization to take into account setup uncertainties and ensure appropriate coverage at the skin. A 5 mm isocenter symmetric shift was used to evaluate robustness. Alliance A221505 trial constraints were used to evaluate the plan. Results: Total PTV coverage achieved was V95%=98.8%, while mean heart dose was kept at 408 cGy, total lung V18Gy=17.7% and V4.8Gy were 63.0%. All Alliance A221505 hypofractionated dose constraints were met at the preferred or acceptable level while highly exceeding all coverage constraints. Robustness analysis showed a worst case scenario of PTV coverage of V95%=94.0%. Minimal breath-hold coaching was needed from the therapy team with the use of patient visual feedback. Each arc was completed in nearly a single breath-hold and treatment was completed in 30 minutes. Conclusion: A dual isocenter DIBH VMAT technique was developed using visual-guidance for comprehensive chest-wall and locoregional node irradiation. The technique was able to meet relevant OAR constraints and an efficient and robust treatment delivery. (2) Not provided, (3) Not provided, (4) Robertwood Johnson Hospital/cancer Institute of New Jersey Purpose: To calculate all the required parameters for avoiding necrosis using quality metrics in order to restrict 12 Gy volume below 10cc. Methods: The dose is generally determined by the function of volume and maximum diameter of the metastatic lesion rather than the histology alone. The quality metric parameters CI, SI, GI are used to calculate normal brain volume more than 12 Gy. The fall off percentage dose is used to develop analytical formula to avoid necrosis. Results: Fractionation can be given according to tumor volume using (1+Quotient(Vol(cc)/4)). The analytical fit of fractionated normal brain 12 Gy equivalent with value of 3 Gy is 12*x0.44, where x is fractionation The accuracy is within 0.5Gy with respect to LQ model. TV*((GI/sel)-1) is normal brain volume of half of prescribed dose, where TV~=TTV, where TV is Tumor Volume, TTV is Treated Tumor, GI is gradient index, sel is selectivity. Volume Accuracy calculated-Measured in %/% 7+-6, Where TV tumor volume, GI-gradient Index, sel-selectivity.Analysis from the isodose fall off resulting from the arrangement of filling cones a. Large:small as 50:50 during optimizationThe analytical formula for TMR10 algorithm is fit in the form A.k-x for calculating any required iso volume outside the tumor, where A is initial volume, k is Required iso percentage and x is constant -1.5. Similarly done for the arrangements with size of the cones 80:20, 20:80.The initial value A is a function of tumor volume given byA= 450*PIV-450, Provided Gradient Index<3, where PIV(PIV=TV/sel) is prescribed iso volume in cc, given by TV/sel. A is representative of Lower IsoVolume nearing zero IsoVolume. Experimentally max of 5% difference is obtained and selectivity is less than 0.7. Conclusion: The above mentioned analytical formulas can be used for quick check on parameters and protecting normal brain. Purpose: Total or partial scalp irradiation is used to treat variety of scalp lesions including squamous cell carcinoma. The close proximity of brain and other critical structures make electron therapy and high dose rate brachytherapy (HDR) desirable methods. The delivery of these methods remain challenging due to the complex shape and curvature of the scalp. The attachment of HDR skin applicator is of particular importance due to need for homogenous dose distribution throughout the target area. Clinically common techniques are often unreliable and can result in partial misdelivery of treatment. Here, we propose to use swimmers cap to help ensure accurate contact of applicator to skin and homogenous treatment delivery. Methods: Freiburg flap was used as the skin applicator and cut to size to match the target. Adhesive tape was used to place the flap onto the scalp. Later the swimmer's cap was placed covering all of patient's scalp. It was apparent that the flap was in contact with the skin tightly. Results: Patient's CT scan revealed that the placement of the flap was impeccable and all of the edges were in contact with patient's skin. An HDR plan was generated later and homogenous dose distribution was obtained throughout the target. 100% of the target area was getting 100% of prescribed dose while %50 isodose line was about a maximum of 2.3 cm away from the surface. We found the placement of the flap much easier and reproducible than existing methods. Conclusion: Accurate placement of the flap is crucially important in terms of accurate treatment delivery. The use of swimmers cap was easy to place and reproducibility was reliable. The patient had minimal to no discomfort. Swimmers cap is proven to be an easy and beneficial method to be used during skin HDR on scalp. Purpose: Radiation therapy workflow encounters challenges in daily patient repositioning. Foam immobilization molds were a developed solution in collaboration with .decimalTM (Sanford, FL) at our facility. This addressed a variety of treatment modalities including proton therapy, MRI guided therapy, SBRT as well as clinical indications for routinely difficult setups such as extremity treatment. It was also used to exact patient treatment positions to PET scan positions for better targeting of metastases. This presentation details the clinical indications for this device, its use, and its accuracy. Methods: To date, foam immobilizers have been applied to 15 different cases. First, a patient body contour from volumetric imaging was transferred to .decimalTM for foam manufacture. For each device, 3D volumetric scans of the patient with and without the foam molds were compared for fit to patient anatomy as determined by rotational corrections from image fusion using Eclipse treatment planning system (Varian Palo Alto, CA). For fractionated treatment courses, the daily setup standard deviations were evaluated to check interfraction setup variations using the foam immobilization device. Results: The .decimalTM foam immobilizers provided excellent fit, with low average pitch 1.19° ± 1.24°, yaw 1.01° ± 1.19°, and roll 0.95° ± 1.20°. There was consistent daily setup variability, with a mean rectangular shift of 3.90 mm ± 3.32 mm and a mean angular shift of 0.56° ± 0.70°. Rectangular shifts may have been larger due to unusual cases that were offset at initial setup for imaging purposes. Conclusion: We have developed a method for creating patient specific foam immobilization that can be used to replicate the patient diagnostic scan position or can be used to immobilize patients in configurations that are not possible with conventional vacuum molds. This process was used on several cases and can be implemented to any treatment site or treatment modality. TX, (2) Methodist Hospital, Houston, TX, (3) Methodist Hospital, Houston, TX, (4) Houston Methodist Hospital, Houston, TX Purpose: 30-fraction fractionated stereotactic radiotherapy is the standard of care in our clinic for pituitary tumors due to normal tissue repair as well as for the safer management of treatment setup errors compared with single-fraction stereotactic radiosurgery (SRS). However, the conformal dose provided by SRS can prove helpful in avoiding the several organs at risk near the pituitary. Planning these cases with HyperArc (HA) also provides short delivery times and avoids potential collision issues. Our goal is to combine these benefits by planning pituitary cases with HyperArc while still delivering multiple fractions. Methods: Five pituitary patients treated with 30-fraction VMAT were replanned using HyperArc on Eclipse v15.6. The optimization objectives were unchanged, but the SRS NTO optimization objective was used. Conformity index, Paddick conformity index, gradient index, and heterogeneity index are evaluated and compared. Results: The average conformity index (CI) is 0.984±0.019 for HA and 1.044±0.029 for VMAT. Average Paddick conformity index (PCI) is 0.871±0.011 for HA and 0.838±0.029 for VMAT. Average gradient index (GI) is 2.824±0.216 for HA and 3.739±0.326 for VMAT. Average heterogeneity index (HI) is 1.285±0.103 for HA and 1.216±0.078 for VMAT. Two-tail t-test shows that the differences in the averages are statistically significant (α=0.05) for CI and GI, but not for PCI and HI. Conclusion: On average, PCIs and HIs are comparable between VMAT and HA plans. CI and GI improvements indicate potential superior sparing of organs at risk. HI is somewhat larger for HA than VMAT plans, however the maximum doses are still reasonable. IMRT QA using portal dosimetry confirmed there are no significant delivery issues with the HyperArc plans. , CARTI, Little Rock, AR Purpose: In version 15.5, Eclipse TPS from Varian introduced a new option in the PhotonOptimizer (PO) called Aperture Shape Controller (ASC). The ASC can increase the MLC opening area and decrease the MLC modulation complexity. Stereotactic radiosurgery treating multiple brain lesions using a single isocenter plans can be complex due to MLC movement restrictions and fewer degrees of freedom when optimizing. The aim of this study is to evaluate the difference in plan quality using the ASC during optimization. Methods: Ten previously treated patients planned using Varian Eclipse version 13.6 were identified. All of these patients were treated using a single isocenter technique targeting 2 to 5 lesions. Each of the plans were then re-optimized using PO version 15.6, with the ASC set at none, moderate, and very high, with a total of 30 plans generated. The number of arcs, gantry angles, and collimator rotations were kept the same for each patient. After optimization, each plan is renormalized so that each PTV is achieving V100% greater than 99% of prescription dose. For each set of plans, the difference in MU, gradient measure, conformity index, and theV10Gy of the brain-minus-PTV structure is calculated and compared. Results: The gradient measure and conformity index were comparable between each set of plans.The MU for the moderate ASC setting plan on average is slightly reduced compared the no ASC plan. The MU for the very high ASC setting plan is slightly higher than the no ASC plan. The brain-minus-PTV V10Gy of the moderate and very high ASC plan increased slightly compared to the no ASC plan. Conclusion: ASC enabled in optimization can achieve similar conformity and gradient index compared to regular plans. ASC can be an effective tool to decrease the plan MU and the complexity of SRS plans. , (1) (2) Radiological Imaging Technology, Inc., Colorado Springs, CO, (3) Mass General/North Shore, Danvers, MA Purpose: To evaluate the historical quality of a stereotactic treatment machine, and its environment, using a comprehensive isocenter optimization routine developed by Radiological Imaging Technology, Inc. (RIT). Methods: In collaboration with Massachusetts General Hospital (MGH), five years of historical data was accumulated within a RITtrend® database. Data was collected daily for five field images, and monthly for twelve field images. In this way, 531 records were collected between 2015 and 2019 and monitored for relevant trends and specifications. The data was referenced against the 3D Maximum Measured Deviation at a nominal tolerance of 1 mm, according to AAPM TG-142 guidelines. For daily stereotactic QA, five fields were taken at the following angle configurations: (gantry = 0°, table = 0°), (g=90°, t=0°), (g=135°, t=0°), (g=230°, t=40°), and (g=330°, t=340°). These angles were deemed significant to patient treatment. For the monthly comprehensive optimization, twelve fields were taken at the recommended angle configuration provided in the RIT manual. Results: The 3D displacement showed a rhythmic sinusoidal pattern with a mean of 0.4 mm and a mean period of 6.2 (S.D. 1.7) months. A distinct shift of about 0.6 mm in the z-axis component of the 3D displacement was also observed over the span of several days. Statistical control limits imposed on the dataset clearly alerted to numerous measurements that, though were generally within the accepted tolerance, fell out specification for the machine variance one might expect from a well-behaved system. Conclusion: This work demonstrates the strength of monitoring machine QA with a historical perspective. The context of the data over an extended period elucidated several potentially alarming trends that may had otherwise gone unnoticed upon individual inspection of the measurements. With regards to the quality of the treatment machine and its environment, we observe this methodology's high importance to overall machine QA procedures. Purpose: Skin tattoos represent the long-held standard for surface alignment and setup of breast cancer radiotherapy yet contribute to adverse cosmesis and patient dissatisfaction. With the advent of contemporary surface imaging technology, we evaluated setup accuracy, setup time, total in-room time, and dosimetry between "tattoo-less" and traditional tattoo-based setup techniques. Methods: Patients receiving accelerated partial breast irradiation (APBI) underwent traditional tattoo-based setup, alternating daily with setup by AlignRTv5.1 (tattooless). Following initial setup, the position was verified by daily kV imaging, with matching on surgical clips to represent ground truth. Translational shifts prompted by kV imaging were ascertained as were setup time and total in-room time. Delivered dosimetry was calculated using the reverse isocenter shift technique. Statistical analysis was performed using Paired Samples t-test. Results: For the first ten patients analyzed, the mean shifts were -0.215±0.146 cm, 0.111±0.138 cm, and -0.148±0.174 cm for the AlignRT-based setup in vertical, lateral, and longitudinal direction respectively. The mean shifts using tattoo-based setup were 0.095±0.181 cm, -0.06±0.326 cm, and 0.075±0.105 cm in vertical, lateral and longitudinal direction respectively. AlignRT-based setup was not significantly different from tattoo-based setup in terms of translational shifts. No statistically significant difference was found in patient setup time and total in-room time between the two techniques. However, AlignRT-based setup showed a narrower distribution in setup time with fewer lengthy outliers as in the tattoo-based approach. The setup time range for tattoos was 157-2481 seconds versus for AlignRT: 215-1323 seconds. Conclusion: These results suggest that a tattooless setup approach to breast radiotherapy with AlignRT may be sufficiently accurate to obviate the need for surface tattoos. Further analyses among large cohorts and including all six degrees of freedom (three translational dimensions along with pitch, roll, and yaw) will determine whether tattoo-based approaches can be replaced by surface imaging. Purpose: The aim of this study is to evaluate and compare hippocampal sparing whole-brain radiation therapy treatment planning techniques using volumetric modulated arc therapy and intensity modulated radiation therapy (IMRT) Methods: 12 patients previously treated with whole brain radiotherapy, re planned with Volumetric Modulated Arc Therapy and Intensity Modulated Arc Therapy; in an effort to spare the hippocampus region. To delineate optic chiasm, optic nerve; magnetic resonance imaging (MRI) and computed tomography (CT) data acquisition and their fusion has been done. The hippocampus avoidance region was created by giving 5 mm margin to the hippocampus. RTOG 0933 recommendations were applied for the beam arrangement purpose and treatment planning. The prescription for treatment was 30 Gy in 10 fractions and the planning PTV was drawn incorporating the while brain including the hippocampus region. Results: Dose to hippocampus by IMRT (DMLC) and VMAT are 1786.78±12.53 cGy Purpose: Aim of this study is to evaluate the compatibility of robotic couch assisted six-dimensional (6D) patient positioning systems with external independent imaging modalities for two different accelerators, Varian and Elekta. Methods: In both machines integrated image guidance systems like cone-beam CT (CBCT) provide patient set-up errors that can be corrected by the respective 6-D couch. Additionally, in-room image guidance systems (BrainLAB's ExacTrac) can be integrated into the workflow to provide pre-, post, and intrafraction set-up errors. Varian 6 degrees of freedom (DoF) couch, comprises three translational axes (x, y, z), table column's rotational axis (yaw). Pitch and roll rotations are done by the couch's bellows. After 6-D correction, couch remains physically rotated about the vertical axis (yaw: zROT≠ 0). In Elekta's HexaPOD 6 DoF couch, the column rotational axis is not in the 6 DoF. HexaPOD's bellows can execute all three rotations (pitch, roll, and yaw) and the 3D couch movements are limited to the translational axes. All positional corrections are isocenter focussed. Results: Varian's way of correcting the rotational error in yaw direction makes it partially incompatible with in-room ExacTrac. In case patient's 6D setup correction is initiated with delta shifts obtained from in-room imaging, the resultant finite zROT inhibits incorporation of CBCT in the subsequent workflow. On the other hand, if the patient's 6-D setup correction is initiated from Varian's CBCT, the resultant zROT in subsequent stereoscopic in-room imaging would reflect as a deviation from planned rotation values. However, in Elekta, since zROT occurs only for the 6-D couch top without involving any column rotation, it is possible to flip between or combine the CBCT and the in-room imaging any time during the workflow. Conclusion: One of the commercially available 6-D robotic couch solutions (Varian) is only partially compatible with the independent in-room imaging system. Purpose: Radiation treatment to the left breast is associated with increased cardiac morbidity and mortality. The deep inspiration breath-hold technique (DIBH) can decrease the radiation dose delivered to the heart and lung. Deep Inspiration Breath-hold, utilizing a respiration-monitoring device by using an LPT system has been used in our clinic to reduce the cardiac dose and lung dose for patients receiving left-sided breast irradiation compared to freebreathing (FB). Methods: Between July and October 2015, a total of 10 patients with left-sided breast cancer underwent two computed tomography scans each with the DIBH using the LPT system and using FB after mastectomy. The scans were retrospectively re-planned using standardized criteria for the purpose of this study. Treatment plans were generated by the 3DCRT technique. The DIBH plans for each patient were compared with FB plans using dosimetric parameters. Results: All patients were successfully planned with the DIBH technique using the LPT system. Significant differences were found between the DIBH and FB plans for mean heart dose (4.49 vs. 5.95Gy,p=0.009), heart V30 (4.74 vs.6.82 %,p=0.006), V20 (6.41vs.9.12 %,p=0.004),and mean left anterior descending coronary artery (LAD) dose (20.39 vs. 24.93Gy, p<0.001).The mean left lung dose (8.18 vs. 9.29Gy, p=0.001) and lung V20 (20.26 vs. 21.82%, p=0.1366). There was no significant difference in the mean left lung dose, mean right lung or mean dose to the contralateral breast. Conclusion: This study reports that the use of a DIBH technique using LPT system in breathing adapted radiotherapy for leftsided breast cancer is easily feasible in daily practice and significantly reduces the radiation doses to the heart, LAD and lung, therefore potentially reducing cardiac risk. Purpose: To evaluate the ability of the Veneziaᵀᴹ advanced multi-channel tandem and ring applicator to consistently produce dosimetrically comparable plans utilizing a reduced number of needle channels, to reduce the risk of secondary complications when boosting cervical cancer treatments with HDR brachytherapy. Methods: We evaluated 26 fractions from 13 patients who were treated with HDR brachytherapy using the Veneziaᵀᴹ applicator. The original plans included a full load of 12-16 needles, including both parallel and 30⁰ oblique needles. We replanned each original to 9 new configurations, each with a reduced number of needles. Comparisons included differences in percentage dose coverage to 90% of the high-risk CTV (HRCTV), and percentage dose to 2 cm³ of the bladder, rectum, sigmoid, and bowel. We considered new plans "passing" if they remained within our standards (D90 > 100%, D2 cm³ < 85% bladder, < 65% rectum, sigmoid, bowel), or did not perform worse than original. Results: Removing only the two most anterior needles showed a 80.8% passing rate. Removing only the most posterior two needles from both sides, or the most anterior and posterior four needles together, both showed a 65.4% passing rate. Removing all oblique needles showed a 19.2% passing rate. Removing only left-sided or only right-sided oblique needles showed 46.2% and 23.1% passing, respectively. Removing only right-sided or only left-sided parallel needles separately showed 19.2% and 34.6% passing, respectively. Removing all parallel needles showed a 11.5% passing rate. Conclusion: As only two replans required a full needle load to maintain dosimetric quality and 40, 36, 18, and 10 replans passed with 2, 3, 4, and 6 needles removed respectively, this indicates the potential for using a lesser number of interstitial needles during combined intracavitary and interstitial HDR brachytherapy while maintaining dosimetric quality. Purpose: Collision detection software has been developed and implemented by others, however they require the treatment planning system (TPS) to be running locally. This report incorporates a collision detection software directly into the TPS, making it accessible to on-site and remote users. This software also includes the ability to account for anatomy not included in the CT scan. Methods: A contour of the patient's skin and of the couch exterior are created by the planner. Those structures are then segmented within the software. Isocenter is found for the specific beam that is being tested. Using characteristics of the beam, an arc is created about isocenter with a radius equal to the distance from isocenter to gantry head. The software then searches the segmented structure on each slice for the closest point to the generated arc. The closest distance for the structure is reported back along with the angle of the gantry head at that point. This software was tested against a phantom for plans containing couch rotations and shifts. Results: The software was tested against a Solid Water phantom with various shifts applied. When shifts resulting in collision were applied, the software was able to correctly predict the angle of collision. For plans without collisions, closest approach distances from phantom to gantry head were predicted within 4.5 cm. Conclusion: The collision detection software was incorporated into the Eclipse TPS. This tool has the potential to eliminate replanning by reporting collisions during the planning process, thus improving the quality and timeliness of patient care. With the ability to be used by remote users as well as on-site users, this tool is both accessible and accurate. Purpose: To characterize the radiation response of 3D printed materials for photon, electron, proton, and CT in absolute terms and as compared to a Plastic Water® baseline for extrapolation to clinical use. Methods: Six materials were 3D printed into blocks using a Fused Deposition Modelling printer. Measurements were made using electron beam to create PDD curves, using photon beam to create TMR curves, and using proton beam to determine the RSP of the materials. The materials were also scanned using CT to examine the variations in HU value within each block and between blocks. The effective density of each printed block was determined to examine the variations in the printing. The densities of the block were used to analyze the results of the CT, electron, photon, and proton results. Results: The effective density of each material varied widely between blocks and within each block. All results for each radiation type were dependent on the effective density of the 3D printed material, with an approximately linear relationship with the average HU value of the material, the R50 of the PDD curves for electron, and the RSP of the block using proton beam. Materials with densities lower than Plastic Water exhibited a negative percent difference trend compared with Plastic Water®, and materials with densities higher exhibited a positive trend. Conclusion: Although 3D printing has much promise for use in radiation oncology, establishing a solid quality assurance protocol prior to its implementation is key to an accurate and successful clinical application. It is recommended that each 3D printed object be properly characterized before clinical use, including determining the effective density. Through the implementation of these measures, 3D printing in the clinical setting has the potential to further improve patient care within radiation oncology departments. The University of Oklahoma Health Sciences Center funded the purchase of 3D printing materials. Purpose: VMAT planning provides highly conformal dose distributions and increased delivery speed when compared to 3D techniques. Knowledgebased VMAT planning for comprehensive breast/chest-wall irradiation can spare heart and ipsilateral lung while providing improved coverage to the internal mammary nodes (IMN) without adversely affecting low dose constraints on contralateral lung. Methods: A knowledge-based RapidPlan model was constructed using 155 IMRT comprehensive breast treatment plans. A standard VMAT technique was developed to spare the heart and ipsilateral lung as much as possible without sacrificing coverage to the IMN. The technique is compared against the RTOG 1304 Group 2-Arm2A benchmark case using standard fractionation. The dose distribution for 65 comprehensive breast and chest-wall plans treated in our institution in 2019-2021 using a hypofractionated approach were analyzed for the following dose metrics: V95% of combined PTV, dose homogeneity index (D95%/D5%), RTOG conformity index, and adherence to the Alliance A221505 trial constraints. Results: The VMAT plan technique was able to successfully meet all of the RTOG 1304 Group 2-Arm2A constraints on the benchmark case with a mean heart dose of 2.96Gy, ipsilateral lung V20Gy=22.7%, and V5Gy=6.44%/8.84% for contralateral lung/breast while achieving V95% > 96.7% for all PTVs. The evaluation of the clinical cases showed an average: mean heart dose of 416.3±102.16cGy for LT-sided targets and 286.1±73.64cGy for RT-sided targets, ipsilateral lung V18Gy of 21.8%±5.77%, contralateral lung V4.8Gy of 15.4%±5.37%. Average total PTV coverage of 95% RX was to 98.8%±1.01 of volume with an average conformity and dose heterogeneity of 1.2±0.10 and 0.92±0.02 respectively. Conclusion: VMAT breast planning can consistently help reduce heart and ipsilateral lung doses for comprehensive breast irradiation of the whole breast/chest-wall and locoregional nodes; while achieving highly conformal and homogeneous dose to the target and only a minor and acceptable increase to low dose constraints to the contralateral OARs. Mr. Schmidt reports personal consulting fees and honoraria with Varian Medical Systems outside of the submitted work. Remaining authors have no conflicts to disclose. Q&A What is a project? A project is temporary activity that has defined beginning and end in time, and therefore defined scope and resources. A project is designed to accomplish a singular go, often including people who don't usually work together. Why should physicists learn project management tools? As physicists, we are involved in projects in a variety types of projects yet we have rarely received formal project management training. Project management was last talked about at Leadership Academy in 2016. A review of project management tools would help our physicists: 1. Strengthen communication & collaboration. Healthy communication is instrumental in managing a successful team and completing a project. If your team is struggling with communication, using a few basic project management concepts will aid collaboration. 2. Supports agility. Projects rarely go as planned. Deadlines shift and expectations change. How to prevent project creep and be agile will determine how quickly and efficiently a project will get back on track. 3. Builds confidence. Having the right tools helps build confidence and trust across teams. Physicists who are good project managers become vital to the function of the organization and provide greater visibility of the profession. Learning Objectives: 1. Understand what project management is and its role in operations 2. Understand the physicists role in projects and improve project execution. 3. Learn project management basics/tools to utilize immediately in the work environment. Purpose: A Monte Carlo (MC) beam model and its implementation in a clinical treatment planning system (TPS, Varian Eclipse) are presented for a modified ultra-high dose-rate electron irradiator and FLASH radiotherapy (eFLASH-RT) utilizing clinical accessories and geometry. Methods: GAMOS MC software modelled the gantry head without scattering foils or targets, representative of the LINAC modifications. Applying a nominal and mean energy of 10MeV, the energy spectrum's standard deviation (σₑ) was varied to match the central axis percent depth dose (PDD) profiles of an open field with jaws retracted and measured with Gafchromic film. The beam's source spot size was 0.5mm. The emittance cone angle (θcone) was determined by equating the quadratic fit between simulated θcone and resulting lateral spread to the film measured lateral spread at several depths. The beam model and its Eclipse configuration were validated with film measured lateral profiles in air and water, and PDD's of the open field and clinical applicators. Eclipse produced an oral melanoma treatment for a canine patient as a demonstration. Results: The GAMOS model agreed best with measured profiles at σₑ=0.5MeV and θcone=3.9°±0.2°. The quadratic fit excluded θcone>4.63° due to obstruction from the primary collimator. The model and its Eclipse configuration agreed with measured profiles (3% for in-water lateral, 4% for in-air lateral, and 3% for PDD's). The anterior-posterior treatment can spare the tongue (<10% receiving >9Gy) while delivering >90% of the prescribed dose to majority of the tumor volume. Conclusion: To the best of our knowledge, this is the first functional clinical TPS configured for eFLASH-RT and it can narrow the current gap in FLASH-RT translation from preclinical to clinical implementation. The optimization of dose and dose-rate can determine the quality of eFLASH-RT plans, especially in cases of large animal and human patients involving complex geometries and tissue inhomogeneities. Purpose: This work outlines a robust solution to MR-Linac daily QA, including mechanical-imaging-radiation isocenter coincidence verification using Cherenkov imaging. Methods: A fully enclosed and sealed acrylic cylindrical phantom was designed to be mountable to the manufacturer-provided jig. A custom-machined plastic conical structure was fixed inside the phantom and held in place with 3D-printed spacers. The phantom was filled with water, allowing for high edge contrast of the plastic cone on MR images. Both a star shot plan and a four-angle sheet beam plan were developed and delivered to the phantom; the former allowed for radiation isocenter localization in the x-z plane (A/P and L/R directions) relative to physical landmarks on the phantom, and the latter allowed for the longitudinal position of the sheet beam to be encoded as a ring of Cherenkov radiation emitted from the phantom, allowing for isocenter localization on the y-axis (S/I directions). A custom software application was developed in MATLAB to perform near-real-time analysis of the data. Results: Calibration procedures showed that linearity between longitudinal position and optical ring diameter is high (R² > 0.99), and that RMSE is low (0.184 mm). The star shot analysis showed a minimum circle radius of 0.34 mm. The final isocenter coincidence measurements in the lateral, longitudinal, and vertical directions were -0.61 mm, 0.55 mm, and -0.14 mm respectively, and the resulting 3D coincidence was 0.83 mm, below the 2 mm tolerance defined in TG-142. Longitudinal analysis showed an average coincidence of 1.5 mm ± 0.4 mm over 8 weeks of daily use. Conclusion: This novel work highlights an efficient method for acquisition and near-real-time analysis of MR-Linac isocenter coincidence data and represents a direct measurement of the 3D isocentricity. This phantom and custom analysis application combined with longitudinal analysis makes this solution easily transferrable to clinical use. This work has been funded by NIH grant R01EB023909. Brian Pogue and Petr Bruza are affiliated with DoseOptics LLC, which provided hardware support for this study. Daniel Alexander is a research consultant for DoseOptics LLC outside the context of this work. Purpose: Current MR-Linacs use MR-guided beam gating at fixed delivery angles. Gantry motion introduces significant image artifacts and thereby inhibits real-time target tracking. This work evaluates a novel deep learning-based approach for gantry motion artifact reduction using clinical MR-cine data acquired during routine clinical use of an MR-Linac. Methods: MR-cine image sequences from patients undergoing radiotherapy were accessed under an IRB approved protocol. A quantitative metric to detect gantry motion artifacts was developed and utilized to partition the dataset into artifact and artifact-free image sequences, with two-thirds used for training and one-third for testing. A previously developed deep neural network model (known as cycleGAN) was trained to synthesize artifact-free images from those containing artifacts and vice versa. Model predictions for artifact removal were then assessed on the test dataset. Results: A total of 1,664 cine images from five patients were included in the analysis (train: 653 artifact and artifactfree each, test: 358 artifact). Artifact detection using frame-by-frame MSE was robust metric to partition artifact-free and artifact data in all patients. Using the network model on artifact frames in the test set resulted in an increase in both structured similarity index measure and peak signal-to-noise (paired t-test, SSIM: 0.83 vs 0.72, p<0.01; PSNR: 24.5 vs 21.3, p<0.01). Performance on 135 frames from a patient withheld from training was comparable to average performance (SSIM: 0.83, PSNR: 25.3), indicating robustness against patient specific anatomy. Computation time for processing of a single image was less than 20 milliseconds in deployment. Conclusion: This study is the first to propose a computationally efficient solution for reduction of gantry motion artifacts on MR-Linacs and could help enable real-time target tracking for volumetric modulated arc therapy. Future studies should evaluate the clinical utility of the approach and its ability to support real-time target tracking during gantry motion. SA-B-Therapy Room-04 Computer Vision Mechanical QA: The Future Is Automation R Ger*, M Armstrong, D Robertson, Mayo Clinic Arizona, Phoenix, AZ Purpose : To use computer vision to automate monthly mechanical QA tests on a linac, including table travel, ODI, field size detection, table and collimator angle, and table and collimator walkout. Methods: A 5MP digital camera with an 8.5mm focal length lens was connected to the interface mount of a Varian TrueBeam with a custom holder. A custom board was created with ARUCO markers, which give their location and a unique identifier for each marker, allowing orientation of the calibration pattern to be unambiguously determined. A user-friendly GUI has a button for each test that changes color to indicate passing or failure according to MPPG8a/9a criteria. The GUI also shows the image analysis regions for each test. Reproducibility measurements were taken across four days. Comparisons to ground truth were performed by graph paper or digital level. Twelve physicists were used to set the field size and to read the ODI as no ground truth could be established for these tests. Results: The system takes five minutes to setup and seven minutes to complete all tests. All tests were reproducible within 0.5mm and 0.5°, except for the 40cm field size Y jaws as these have a large light penumbra, and ODI measurements that were reproducible within 0.1cm. Table travel, ODI, and walkout measurements agreed with ground truth within 0.5mm and 0.4°, except vertical table motions that agreed within 0.9mm due to lens focus optimized for the 100cm SSD plane. The average computer vision reading of the field size set by the physicists matched the desired field size within one standard deviation in most cases. Some readings exceeded this value due to the difficulty of precisely locating the broad Y jaw penumbra with the human eye. Conclusion: The computer vision mechanical QA system can reproducibly perform mechanical QA tests with high accuracy. Purpose: To develop a framework for QA assessment and tracking of CBCT-IGRT systems, from routine daily acquisitions. Methods: Daily CBCT scans of a cube phantom (MiMi, Standard Imaging) were retrospectively collected for 6 TrueBeam systems from 3 centers. The number of available scans per machine varied from 138 to 840. CBCT acquisition protocol was constant for each machine; centers used a pelvis protocol except for one (head protocol). A diagnostic CT scan of the phantom was also acquired, as a reference. For each scan, 43 radiomic features were extracted from a 10-by-10 cm ROI positioned in the center area of the phantom's middle slice. Delta radiomics (PyRadiomics) was used to compare features from CBCT data to the reference scan. Service reports were collected to identify whether changes in radiomic features over time correlated with system changes. Results: For a given IGRT system, the radiomic time series showed consistent trends; when a large shift from the baseline was observed for one radiomic feature, corresponding shifts, of varying magnitude, were observed in the other features. Four of the systems showed one or more substantial shifts in their radiomics time series, some of which correlated with events described in the service reports. We identified chiral asymmetries in image quality, due to scan rotational direction, for three systems, a finding that could have been missed by using traditional QA techniques due to its small impact on HU (< 10%). Conclusion: A radiomics-based framework for QA assessment of CBCT-IGRT imaging performance was developed that is fast, automatic, and does not require additional acquisitions to routine QA. Our results suggest that this framework is sensitive to imaging system functionality that could help in early identification of potential malfunctions of the system. Purpose: With the implementation of template-based stereotactic radiosurgery utilizing non-coplanar volumetrically modulated arc therapy (n-VMAT), it is hypothesized that the traditional Winston-Lutz (WL) test may no longer be sufficient to represent isocenter deviation. Triggered MV imaging at specified angular increments throughout a treatment arc could be used to autonomously evaluate isocenter deviation for a given template's geometry. Methods: A 2.5 mm ball-bearing WL phantom was aligned to mechanical isocenter and then kV isocenter to represent a patient setup verification on two Truebeam Linacs. 2 cm MLC defined square aperture images were taken at 5-degree increments for 3 arcs used in a clinical fractionated SRS template and exported to an in-house MATLAB script. 8 conventional static WL images were taken for comparison. For each arc, the isocenter deviation was extracted and plotted as a function of gantry angle and the maximum deviation found in each direction. Potential plan deviations were evaluated by applying the 'worst-case' deviations to each arc in a series of clinically treated SRS plans using identical geometry. Results: The average maximum deviation was found to be larger for the treatment arc WL than shown in the convention WL test (0.75 mm vs 0.49 mm) but still within vendor specification. After applying the maximum isocenter deviations for each arc, target coverage dropped on average for both multi-lesions single-isocenter and single-lesion plans from 99.3 ± 0.45% to 97.7 ± 1.2%, maximum loss of 2.72%. Conformity index for each lesion dropped by 0.023 on average, maximum of 0.0472, whilst there were no statistically significant changes on dose to critical organs including normal brain. Conclusion: We have demonstrated a novel non-coplanar arc-based WL test, via triggered imaging, that better characterizes isocenter wobble for arc-based stereotactic treatments than the traditional static WL test. This method would be specifically useful for single-isocenter/multi-lesions SRS setting. Purpose: To compare radiologist preference for a model-based iterative reconstruction algorithm (FIRST) to a statistical-based iterative reconstruction algorithm (AIDR 3D) using clinical images. Methods: IRB approval was obtained to retrospectively collect and post-process 60 clinical CT exams from lung screening, pulmonary embolism, and non-contrast abdomen protocols (20 from each protocol). These protocols represented three image-quality challenges: low dose, high spatial resolution, and low contrast imaging, respectively. The cases were reconstructed with AIDR 3D and FIRST using default processing settings. Each image pair was presented side-by-side on a diagnostic workstation to three radiologists, with the position of imaging set on left or right randomized. Five image-quality-related categories and four exam-specific clinical categories were rated, with the radiologists indicating a preference strength (strong, slight, or none) and direction (left or right). A two-sided test Wilcoxon signed-rank test was used to analyze the scores in each category, with a p-value < 0.05 considered a significant preference. Analysis was performed for each category and radiologist individually, as well as combined across radiologists. Results: FIRST was not preferred in any category when radiologist scores were combined, and only rarely when separated. For the "overall preference" category, AIDR 3D was preferred for lung screening by two out of three radiologists; the third had no preference. There was no preference between algorithms for pulmonary embolism. AIDR 3D was preferred by one radiologist for non-contrast abdomen, while the two remaining had no preference. FIRST increased breast shield artifacts, but reduced metal implant artifacts. Conclusion: The radiologist image quality study indicated preference for AIDR 3D over FIRST for two out of three imaging quality challenges, with no preference for the third. It is likely that noise texture influenced radiologist preference. Careful clinical implementation of reconstruction algorithm is necessary as FIRST may not be preferred for all clinical applications. Purpose: In this study, we utilized the MC to design a new DMBT VC applicator to address the lack of dose coverage at the VC apex due to anisotropy effect and to minimize the effect of the air gap at the very critical region of the vaginal cuff. Methods: The new DMBT-VC was simulated in GEANT4 MC code, as single lumen and made of PPSU plastic. The central part of that including the lumen was considered to be a detachable 8 mm diameter tandem rod made of PEEK plastic. First, we extracted a standard dose distribution from BrachyVision TPS of the VC applicator with PEEK tandem. Second, to provide directional intensity modulation, the PEEK tandem was replaced with a MR-compatible tungsten alloy tandem with similar dimensions to optimize the coverage at the apex utilizing directional radiation beam generated. Two widely used 192Ir sources, Varian's GammaMedPlus and VS2000 were simulated. Further, we studied the applicator functionality for abating the effect of the apical air gaps. Results: The analysis showed that the new applicator can improve the lack of coverage at the apex due to anisotropy which were about 2mm and 4mm for GammaMedPlus and VS2000, respectively. In addition, the DMBT applicator was able to successfully modulate the prescription isodose line by up to 1 cm at apex without overdosing the peripheral sides through inverse optimization. Conclusion: Results showed the novel DMBT VC can remedy the lack of apex dose coverage due to anisotropy and is very flexible in minimizing the effect of air gaps. This is of major clinical value as the vaginal apex is a frequent site of recurrence in patients. In addition, the applicator can help to potentially reduce dose where to mitigate toxicities (including vaginal mucosa), where anisotropy and airgaps would've been a big barrier in the past. Purpose: to evaluate the utilization of endovascular fiducial markers for real-time image registration and tracking in CyberKnife (CK) (Accuray, Sunnyvale, CA) radiotherapy. Patients at high risk of complications from percutaneous transthoracic fiducial insertion may benefit from an endovascular approach using embolization microcoils. Methods: Three sizes (8x8mm, 2x5mm, 2x3mm) of Platinum Tornado Embolization Microcoils (Cook Inc, Bloomington, IN) were tested for functionality as fiducial markers with the CK real-time tracking system. The detectability of the fiducials on the CK imaging system was first evaluated based on the fiducial tracking algorithm parameters. The 2x3mm microcoil was inserted into a Respiratory Motion Phantom (Modus Medical Devices, Ontario, CAN) to be tested with the institution's 4DCT motion assessment process. To evaluate the microcoil targeting accuracy for planning and treatments, a Synchrony tracking end-to-end (E2E) test was performed, and the total targeting error calculated as the offset between measured and planned 70% isodose lines. Lastly, the patient-specific QA of a clinical lung case was delivered to a MapCheck2 device (Sun Nuclear Corporation, Melbourne, FL), using the microcoils as tracking markers, and resulting dose distribution evaluated using the gamma index analysis with 2%/2mm and 10% dose threshold criteria. Results: Following initial detectability tests, the 2x3mm coil produced fiducial tracking algorithm parameters within manufacturer-recommended tolerances during real-time alignment and is appropriate for clinical use. The lung phantom motion assessment with the microcoils on the 4DCT scan matched the expected breathing motion within 1mm. The total targeting error of the Synchrony E2E test was 0.99mm, which is within the 1.5mm tolerance recommended by AAPM Task Group 135. The patient-specific QA of the clinical lung SBRT plan resulted in a 100% passing rate. Conclusion: The 2x3mm Platinum Tornado Embolization Microcoil performance is satisfactory as a marker for realtime CK tracking and can be employed for CK treatments. Purpose: To evaluate the potential of the high-resolution diode array, SRS MapCHECK (SRSMC), to streamline the commissioning process of HDR brachytherapy surface applicators. Methods: Dosimetric measurements were made for a set of nine brachytherapy surface cones, both with the SRSMC as well as Gafchromic EBT3 radiochromic film at several depths in a solid water phantom. Both phantoms were scanned, and doses were obtained using both the TG-43 formalism and the Acuros BV algorithm with both dose to medium (Dm) and dose to water (Dw) calculated. Comparisons between film and SRSMC measurements were conducted with dose point and profile evaluations at the water equivalent measurement plane of the SRSMC (27 mm). Dosimetric point and profile evaluations and normalized gamma analyses were performed at each relevant depth to compare the measurements and treatment planning system calculations. Results: Examination of profile measurement data between SRSMC and film indicates a strong correspondence at the depth of interest, and the analysis time of the SRSMC is substantially less than film. Normalized gamma analysis comparisons to treatment planning algorithms were above 95% with an evaluation criteria of 2%/2mm for film measurements at 10 mm depth and above 90% for SRSMC at 27 mm water equivalent with a 3%/3mm criteria. Measurements with both systems elucidated a small off-axis irregularity in the dose distribution that results in deviations from the calculated dose profiles. Conclusion: This work indicates the potential to streamline the HDR brachytherapy surface cone commissioning process using the SRSMC. Furthermore, the work conducted provided a thorough investigation of the cone system that will be used to inform quality assurance and clinical implementation. Moving forward, with this work serving as a foundation, clinical experience with surface applicators will pave the transition from TG-43 calculations to Acuros BV. University: Bio-effect Modeling and Combination Therapy with Radiopharmaceutical Therapies SA-C-Therapy Room-04 Q&A Radiopharmaceutical Therapy (RPT) is fast becoming a mainstream modality with the development and approval of new emitters and conjugates. Physicians from both Radiation Oncology and Nuclear Medicine are expressing interest in, learning the fundamentals of, and aiming to implement radiopharmaceutical therapy in their clinic as evidenced by the CE sessions, symposia and training programs organized by ASTRO, SNMMI and Radiation Research (inter alia). However, the efficacy of RPT depends upon delivering a lethal level of radiation to cancer cells while sparing normal cells, and is thus far less susceptible to the resistance mechanisms or inadvertent side-effects that have been observed with biologic (or pathway inhibition therapy) and chemotherapy. To date, the potential of RPT has not been fully exploited. For the most part, treatment has been implemented without using the innate potential advantage of RPT over other treatment modalities -the ability to individualize treatment according to tumor and normal organ absorbed doses and other biologically-driven dosimetric quantities. The dosimetry methodology initially implemented for RPT was developed to assess radiation risk for diagnostic imaging agents. Use of this formalism, along with poor quantitative input, the near complete absence of standardization, not surprisingly, failed to show a compelling benefit to dosimetry-based treatment planning for RPT. Dosimetry formalisms for RPT that provide dose estimates and dose-volume histograms to individual patients based on their anatomy and the biodistribution of the RPT are now available as are substantially improved quantitative imaging methods. This session aims to educate physicists by familiarizing them with the new and forthcoming documents that provide the resources necessary to educate themselves in the fundamentals of RPT dosimetry, including best practices, uncertainty estimation, appropriate methodologies and standardizations of nomenclature and practice. These have been produced by world experts in the field and will provide guidance for a new generation of RPT physicists. These new documents are: 1. the new MIRD 2020 Primer, introduced with a focus on basic bio-effect modeling and the MIRD schema sections; 2. The ICRU Report 31, where the focus of the presentation will be the new nomenclature for RPT, stratified best practice recommendations and indications for combination therapies. 3. The IAEA report, with a focus on Uncertainty, Metrology and Activity Quantification. Learning Objectives: 1. Describe the Medical Internal Radiation Dose (MIRD) and International Commission on Radiation Units and Measurements (ICRU) dosimetry methodologies. 2. Explore methods used in clinical practice to carry out patient specific RPT dosimetry. 3. Understand how to minimize errors and uncertainty in activity quantification in RPT. I, Mammo: Artificial Intelligence in Mammography Q&A Adam Yala -Machine Learning for Healthcare In the first half of the talk, we will discuss how to leverage machine learning to automate tasks humans can do, including the structuring of clinical outcomes from pathology reports and the assessment of breast density. We will use this as an opportunity to introduce the fundamental concepts of supervised learning and to introduce both classical methods such as logistic regression and deep learning. Next, we will discuss how to leverage machine learning to tackle tasks that humans cannot easily do, such as risk assessment and designing dynamic screening policies. In doing so, we will discuss both advanced algorithmic techniques such adversarial training and multi-objective reinforcement learning, and well as the general challenges in developing such systems. Constance Lehman, MD -AI in Mammography: A Physician's Perspective The tools of AI offer promise to tackle some of our greatest challenges in healthcare overall and in mammography specifically. While the initial hype that AI systems will eliminate the role of radiologists in healthcare has waned, questions remain how these tools will be incorporated into "real world" clinical practice. Careful validation studies are needed to determine the true benefit of these tools to avoid a repeat of the history of conventional CAD. As the field evolves to incorporate this new technology into clinical practice, the day-to-day working lives of radiologists will change. This presentation will highlight specific AI tools available for clinical implementation and provide a roadmap for the future of AI in the clinical setting. Specific mammography applications and published research in the specific domains of density assessment, image interpretation, case triage and risk assessment will be reviewed. Q&A In the modern digital age, the sheer amount of patient data available to the Radiation Oncology Team is staggering. Effective management of both patient data and treatment tasks distinguish the successful clinicians from the merely adequate. Towards this goal, the vendors participating in this session have all crafted solutions to meet the exacting needs of the modern Radiation Oncology Team on such fronts as treatment planning, contouring, and quality assurance. Additionally, while there is a great promise of improved efficiencies and treatment outcomes, important topics of discussion will also outline areas of critical review for the supervising clinician, as it is imperative to remember that automation does not absolve the user of responsibility of the output. The purpose of this session is to allow vendors to present their solutions and describe how their product can increase efficiency, plan quality control and enable more extensive adaptive planning processes. Attendees of this session will gain a working insight and knowledge of several vendor's approaches to automating critical radiation oncology workflows. Learning Objectives: 1. See a variety of automated workflow solutions for use in Radiation Oncology 2. Learn where these automations can most effectively be deployed 3. Learn what quality oversight should still be maintained when utilizing these workflows. Q&A Automation is the key to unlocking major progress in radiation therapy. Consistently delivering a first treatment within 1-2 days of initial consult, hypofractionated treatments for chronic metastatic disease, adaptive radiotherapy: These approaches would likely result in better outcomes but are simply not feasible without automation. While contouring, treatment planning, plan evaluation and reporting, analytics, quality assurance, etc. can all be automated to one degree or another, no solution currently exists that does not require clinical expertise at key points to drive the process. Therefore, the goal of automation is to empower clinicians to provide their input in a way that achieves the highest quality results in the least amount of time and with the smallest probability of error. This requires a holistic view into how the automation fits into the broader workflow. Clinical physicists are uniquely positioned to guide their clinics to identify the best candidates for automation and to implement intelligent automation into the clinic's workflow, for example, AI-driven automatic contouring. However, when evaluating commercially-available tools, it is critical for clinicians to understand the hurdles involved in the development and implementation of automation. In addition to commercially-available automation options, many institutions have been developing in-house automated radiation treatment planning tools as a way to improve clinical efficiency. Clinical implementation of in-house tools requires a team effort and assurance of safety is crucial for patient care. Breast treatment planning using tangential beams with fluence maps has been a tedious manual process in our clinic. A machine learning-based automated treatment planning (MLAP) tool was developed to improve planning efficiency for breast treatment planning. After a promising validation study performed both in the research setting and clinical setting, the MLAP tool was launched for actual patient treatment. Performance evaluation in a clinical setting as well as in a prospective fashion is critical to identify strengths and weaknesses of MLAP to enhance MLAP and its future clinical translation. A unified quality assurance guideline also was established to assure safe translation of the in-house automation tool. This guideline covers acceptance, commissioning, and maintenance testing as well as quality assurance test for each upgrade. This session will provide an overview of how to identify appropriate tasks for automation, as well as challenges that may present themselves while trying to implement automation. Additionally, this session will present two methods of clinical implementation of AI: a commercially available AI-driven automatic contouring tool and in-house treatment planning automation tools for breast. Learning Objectives: 1. Learn what to consider when identifying tasks that are good candidates for automation and how to implement a solution in a way that fits your clinic. 2. Understand the key obstacles in developing and implementing automation into clinical workflows. 3. To share experience in clinical implementation of automation tools. Q&A This session will consist of two presentations. The first presentation will focus on the current implementation of tomosynthesis systems, technique selection and impact on image quality, clinical utilization, and dosimetry with some discussion of future trends in clinical and technical development of tomosynthesis techniques. In addition, the impact of image reconstruction technique on image quality will be addressed. With tomosynthesis being applied to many different anatomical exams, understanding of dosimetry techniques and typical dose levels in comparison to projection radiography will be discussed. The second presentation will focus on CT dosimetry. Current clinical CT dosimetry metrics have substantial limitations, which led to the development of Size Specific Dose Estimates (SSDEs) that account for the effects of patient size. The presentation will describe several key elements related to SSDE including: (a) Water Equivalent Diameter, as described in AAPM Report 220, (b) SSDE for body originally described in AAPM Report 204 and updated with AAPM report 220 and (c) SSDE for head described in AAPM Report 293. This presentation will focus on how these metrics were developed, how they are expected to be calculated and how they might be used clinically in the future. Learning Objectives: 1. Understand the impact of radiographic tomosynthesis acquisition and reconstruction parameters on image quality. 2. Describe the dosimetric considerations of radiographic tomosynthesis and opportunities for optimization and clinical testing. 3. Review current radiation dose index metrics used in CT Dosimetry and their limitations. 4. Explain how the Size Specific Dose Estimates (SSDEs) for Body and Head overcome some of these limitations and how they can be used in clinical practice.
doi:10.1002/acm2.13289 pmid:34002941 fatcat:6oivewssf5gwnnztmhff4nldle