Search

Purpose: To investigate whether rotational IMRT is more susceptible to organ motion than IMRT that uses fixed fields. Methods: RapidArc and sliding window IMRT treatment plans were each created for five lung and five pancreas cases. Dosimetric effects due to tumor motion were simulatedby incorporating sinusoidal couch motion into plans using in‐house Monte Carlo dose verification software. Motion was emulated longitudinally (10 mm) for all cases with additional lateral and vertical motion (5 mm) for pancreatic tumors. Dose distributions of original and modified plans were compared using Gamma analysis and dose profiles. Results: Preliminary results show that for one lung case, a 95% Gamma pass rate can be achieved using 6%‐6‐mm criterion for RapidArc versus 8%‐8‐mm for sliding window. Findings are supported by expanding on a well‐known model that demonstrates the effects of intra‐fraction motion on the delivery of dynamic intensity modulation. Conclusions: Rotational IMRT deliveries are believed to be no more susceptible to organ motion than sliding window IMRT, and results thus far support this conclusion.

Purpose: Traditional SBRT employs approximately 10 static beams with up to 20 Gy per fraction, requiring lengthy treatments which can be difficult for patients to tolerate, increasing the risk of movement, and causing discrepancies in the reproducibility of the breathing cycle. Commercial VMAT systems offer shorter treatment times with modulated beams; however, modulation is often not necessary or desired for small fields. Conformai arc therapy offers efficient beam delivery, but with only one aperture shape and constant beam weighting over all gantry angles. This study evaluates the efficiency of a new SBRTdeliveryMethod: a conformai arc with multiple aperture shapes and variable dose rate. Methods: Three clinical SBRT cases were chosen for this study. Each static field was converted into an arc segment to create a StereoArc plan. Gantry angle ranges were determined from the clinical monitor units, with the MU/degree chosen to maximize the dose rate. All segments were merged into a single arc with variable dose rate. Dose distributions from the StereoArc plans were compared to the clinical static field plans using Pinnacle. Delivery times were compared between the static SBRT plans, both with and without Beam Automation, and equivalent StereoArc plans. All plans were delivered on a Varian TrueBeam using a dose rate of 1000 MU/min. Results:Dose differences between StereoArc and static plans were minimal. Delivery times for the static plans were 5–8 minutes, while delivery time with StereoArc was less than 3 minutes for all cases, which was equivalent to delivering the static plans with Beam Automation. Conclusions:Delivery efficiency was improved up to 60%: from 8 minutes for static fields, to less than 3 minutes for StereoArc. StereoArc appears to be both an effective and efficient way of deliveringSBRT for centers not wishing to modulate SBRT and without access to Beam Automation. This study is partially supported by NIH grant 1R01CA133539‐01A2.

Purpose: To evaluate EPID for dosimetry applications of arc and static‐gantry IMRT with sliding window (SW) and/or step‐and‐shoot (SS) deliveries Methods:IMRT beams (SW & SS) were designed that generate beam hold‐offs and dose rate modulation due to MLC motion under 10 × 10 cm jaw. An arcIMRT beam was designed by adding gantry movement to the SW field. A 10 cm × 10 cm open beam was also used. Despite differences in delivered dose rates/pulse characteristics, the four beams should deliver the same total dose. For each beam, various MUs with 6 MV beam at 300MU/min were irradiated on EPID which operated in image acquisition of integration mode (IM), continuous scanning mode with synchronization (CMs) and without (CMn) to beam pulses. Acquired images were evaluated in repeatability, dose linearity, and reproducibility (reproduce open beam output in IM). Results: In IM, repeatability, dose linearity, and reproducibility were within 1% for all dose levels and beams. In CMs, they were within 1–2% if dose rate was maintained steady (1) for SW beam (needed a minimum 1.3 MU/cm MLC motion) and (2) arcIMRT beam (needed a minimum 1 MU/degree and 2.8 MU/cm MLC motion) and (3) if a minimum of 38‐40 MU per shoot was used for SS beam. Nonlinearity was observed for fewer MUs. This is due to the response of EPID to pulse‐length reduction for fixed‐gantry therapy and pulse dropping for arc therapy. The latter produces in‐planar non‐uniformity making EPID unsuitable for temporal dosimetry of arcIMRT. Sacrifice in temporal resolution then became necessary such as multi‐frames per image (eg. ∼1 sec/image). In CMn the results were similar to those of CMs. However, they showed artifacts, thus this mode was not preferred. Conclusions: We found conditions under which integrating and temporal EPIDdosimetry can be used for IMRT and arcIMRT dose deliveries.

Purpose: There is no rigid prescription recommended for the periodic QA of HDR applicators in terms of what must be done and in what frequency. This study investigates the necessity of an annual QA program for HDR applicators, which was performed at an HDR clinic. Methods: In this clinic, all HDR applicators and dummy sources used for HDR treatment are tested on an annual basis. 10 different types of gyn applicators were involved including CT-compatible and non-CT compatible tandem & ovoid set, interstitial tandem &ring set, cylinder set Miami set, vaginal applicator set, ring set, Fletcher Williamson set. The annually QA tests for these applicators included (1) visual inspection, (2) mechanical integrity, (3) source positioning accuracy tests using kV imaging, or a combination of (1)– (3). The source positioning accuracy tolerance was 2mm. Over 100 tests were performed on 48 and 61 pieces of applicator components in 2013 and 2014, respectively. Results: Due to the annual QA, the degraded of applicators or dummy source were ascertained. The degradation either affected the feasibility of source delivery or introduced significant source position errors. The failure rate of the HDR applicator annual QA was 1.6% in 2013 and 2.4% in 2014. Each year, different problems on difference components were detected. The unacceptable applicator components were discarded and replaced by the new orders before affecting any patient treatment. This annual QA takes physicist half to one day to finish depending on his or her experience and expertise. Conclusion: The annual QA tests are determined as essential in clinic and should include visual inspection, mechanical integrity and source positioning accuracy tests. It is strongly suggested to perform annual HDR applicator QA in clinic to increase the confidence to deliver a safe, accurate and appropriate HDR treatment.

Purpose: A method of four dimensional dose reconstruction (4D-DR) using a cine mode of an EPID to determine the delivered 4D dose distribution has been suggested. This method, however, has not been tested under irregular breathing. This study investigated their effects on 4D-DR. Methods: The 4D-DR attempts to find the patient’s breathing phase associated with each EPID image by comparing it to pre-generated-EPID-quality DRRs of every breathing phases. A lung phantom with a tumor object (3cm diameter cylinder) on a moving platform was used for this test under conditions of amplitude reduction by 1/2 during (1) the entire delivery and (2) the duration when MLC blocked the tumor (hindering the phase determination). The dose delivered to the phantom was inversely reconstructed in the associated phase image of the phantom from each EPID image based on our documented inverse method of the 4D-DR; the phase-specific dose was integrated, generating the 4D reconstructed dose. Forward 4D Monte Carlo calculations were used as the 4D dose of a ground truth. Results: The reconstructed dose showed 98.1% gamma pass rate for 3%/3mm under regular breathing (i.e. no amplitude reduction) when compared with the forward 4D dose. When the irregular condition (#1) was adopted for treatment, the pass rate became 94.2%. The irregularity under the MLC block (#2) showed the pass rate of 91.3%. The two conditions did not affect the reconstruction noticeably in DVH plots of the tumor and lung-tumor. Conclusion: As long as the extent of on-treatment motion is smaller than that of the planning CT, the 4D-DR can be done accurately. Otherwise, ontreatment 4D CT images are necessary.

Purpose: To evaluate the role of 2D kilovoltage (kV) imaging to complement cone beam CT (CBCT) imaging in a shift threshold based image guided radiation therapy (IGRT) strategy for conventional lung radiotherapy. Methods: A retrospective study was conducted by analyzing IGRT couch shift trends for 15 patients that received lung radiation therapy to evaluate the benefit of performing orthogonal kV imaging prior to CBCT imaging. Herein, a shift threshold based IGRT protocol was applied, which would mandate additional CBCT verification if the applied patient shifts exceeded 3 mm to avoid intraobserver variability in CBCT registration and to confirm table shifts. For each patient, two IGRT strategies: kV + CBCT and CBCT alone, were compared and the recorded patient shifts were categorized into whether additional CBCT acquisition would have been mandated or not. The effectiveness of either strategy was gauged by the likelihood of needing additional CBCT imaging for accurate patient set-up. Results: The use of CBCT alone was 6 times more likely to require an additional CBCT than KV+CBCT, for a 3 mm shift threshold (88% vs 14%). The likelihood of additional CBCT verification generally increased with lower shift thresholds, and was significantly lower when kV+CBCT was used (7% with 5 mm shift threshold, 36% with 2 mm threshold), than with CBCT alone (61% with 5 mm shift threshold, 97% with 2 mm threshold). With CBCT alone, treatment time increased by 2.2 min and dose increased by 1.9 cGy per fraction on average due to additional CBCT with a 3mm shift threshold. Conclusion: The benefit of kV imaging to screen for gross misalignments led to more accurate CBCT based patient localization compared with using CBCT alone. The subsequently reduced need for additional CBCT verification will minimize treatment time and result in less overall patient imaging dose.

Purpose: A dedicated stereotactic breast radiotherapy device, GammaPod, was developed to treat early stage breast cancer. The first clinical unit was installed and commissioned at University of Maryland. We report our methodology of absolute dosimetry in multiple calibration conditions and dosimetric verifications of treatment plans produced by the system. Methods: GammaPod unit is comprised of a rotating hemi-spherical source carrier containing 36 Co-60 sources and a concentric tungsten collimator providing beams of 15 and 25 mm. Absolute dose calibration formalism was developed with modifications to AAPM protocols for unique geometry and different calibration medium (acrylic, polyethylene or liquid water). Breast cup-size specific and collimator output factors were measured and verified with respect to Monte-Carlo simulations for single isocenter plans. Multiple isocenter plans were generated for various target size, location and cup-sizes in phantoms and 20 breast cancer patients images. Stereotactic mini-farmer chamber, OSL and TLD detectors as well as radio-chromic films were used for dosimetric measurements. Results: At the time of calibration (1/14/2016), absolute dose rate of the GammaPod was established to be 2.10 Gy/min in acrylic for 25 mm for sources installed in March 2011. Output factor for 15 mm collimator was measured to be 0.950. Absolute dose calibration was independently verified by IROC-Houston with a TLD/Institution ratio of 0.99. Cup size specific output measurements in liquid water for single isocenter were found to be within 3.0% of MC simulations. Point-dose measurements of multiple isocenter treatment plans were found to be within −1.0 ± 1.2 % of treatment planning system while 2-dimensional gamma analysis yielded a pass rate of 97.9 ± 2.2 % using gamma criteria of 3% and 2mm. Conclusion: The first GammaPod treatment unit for breast stereotactic radiotherapy was successfully installed, calibrated and commissioned for patient treatments. An absolute dosimetry and dosimetric verification protocols were successfully created.

Purpose: Calculation-based patient-specific IMRT/VMAT QA, as a DICOM-RT-based plan verification system, has been commissioned and is currently being used at our institution. We report our investigation of IMRT/VMAT plan verification failures/warnings out of more than 1,000 plans verified over an 18-month period. Methods: Seven patients with QA failures/warnings from our calculation-based IMRT/VMAT QA were selected. Dose to 95% of PTV volume (D95%) and mean PTV dose with a difference of greater than 3%, a global gamma passing rate of less than 95%, and a mean dose difference to organ at risk (OAR) larger than 5% were set as the failure/warning criteria. The treatment sites of the selected patients included pelvis, leg, head and neck, brain, and breast. Measurement-based IMRT/VMAT QA was done on each patient using MapCHECK2. Root cause analysis (RCA) was performed on all patients to investigate the IMRT/VMAT failures/warnings. Results: All seven patients’ plans passed measurement-based IMRT/VMAT QA with a gamma passing rate of more than 95%. Our RCA of patients showed the IMRT/VMAT QA failures detected by the calculation-based method were mainly due to inclusion of the skin in PTV, and tissue heterogeneity effects particularly at tissue-air and tissue-bone interfaces. Conclusion: Calculation-based IMRT/VMAT QA can detect potential failures which cannot be detected by standard measurement-based QA.

Purpose: This study's intent is to test if the first 3 fractions of an IGRT protocol consisting of a kV orthogonal pair and subsequent CBCT are representative of all remaining treatment fractions. Methods: At our institution the standard IGRT protocol includes CBCT after kV imaging for the first 3 fractions followed by daily kV and weekly CBCT. We randomly selected 20 lung patients who have at least 8 CBCTs in the entire treatment course (total 187 sets of orthogonal kV images plus CBCT). Our evaluation criterion is the fraction of kV-vs-CBCT position mismatches in the first 3 fractions. An additional CBCT is acquired if only 1 fraction mismatches among the first three. We define as 'protocol eligible’ (eGroup) the group of patients for whom the setups with kV images match within the pre-selected thresholds of 3, 4 and 5 mm those of the consecutive CBCTs for at least 60% of the fractions. A protocol ineligible (iGroup) is defined as the group of patients other than eGroup. Results: The fractions of patients in the eGroup were 70%, 85% and 95%, when the setup thresholds selected were 3, 4 and 5mm respectively. The sensitivities of determining iGroup from the first 3 fractions of kV and CBCT were 0.83, 1.0, and 1.0, and the specificities were 0.93, 0.94 and 0.89 with thresholds 3, 4 and 5 mm respectively. Conclusion: The method suggested here well categorized cases for our kV-based on-line IGRT protocol. More patients are eGroup eligible if larger setup threshold is used. CBCT frequency in the on-line daily kV plus weekly CBCT protocol may need to increase for protocol ineligible patients. Also, our IGRT CBCT frequency could be decreased when position differences between kVs and CBCTs are less than 1.5mm, the average positional standard deviation of out kV-CBCT sets.

Purpose: The RavenQA system (LAP Laser, Germany) is a QA device with a phosphor screen detector for performing the QA tasks of TG-142. This study tested if it is feasible to use the system for the patient specific QA of the Volumetric Modulated Arc Therapy (VMAT). Methods: Water equivalent material (5cm) is attached to the front of the detector plate of the RavenQA for dosimetry purpose. Then the plate is attached to the gantry to synchronize the movement between the detector and the gantry. Since the detector moves together with gantry, The 'Reset gantry to 0’ function of the Eclipse planning system (Varian, CA) is used to simulate the measurement situation when calculating dose of the detector plate. The same gantry setup is used when delivering the treatment beam for feasibility test purposes. Cumulative dose is acquired for each arc. The optical scatter component of each captured image from the CCD camera is corrected by deconvolving the 2D spatial invariant optical scatter kernel (OSK). We assume that the OSK is a 2D isotropic point spread function with inverse-squared decrease as a function of radius from the center. Results: Three cases of VMAT plans including head & neck, whole pelvis and abdomen-pelvis are tested. Setup time for measurements was less than 5 minutes. Passing rates of absolute gamma were 99.3, 98.2, 95.9 respectively for 3%/3mm criteria and 96.2, 97.1, 86.4 for 2%/2mm criteria. The abdomen-pelvis field has long treatment fields, 37cm, which are longer than the detector plate (25cm). This plan showed relatively lower passing rate than other plans. Conclusion: An algorithm for IMRT/VMAT verification using the RavenQA has been developed and tested. The model of spatially invariant OSK works well for deconvolution purpose. It is proved that the RavenQA can be used for the patient specific verification of VMAT. This work is funded in part by a Maryland Industrial Partnership Program grant to University of Maryland and to JPLC who owns the Raven technology. John Wong is a co-founder of JPLC.

Purpose: Monitoring the delivered dose is an important task for the adaptive radiotherapy (ART) and for determining time to re-plan. A software tool which enables automatic delivered dose calculation using cone-beam CT (CBCT) has been developed and tested. Methods: The tool consists of four components: a CBCT Colleting Module (CCM), a Plan Registration Moduel (PRM), a Dose Calculation Module (DCM), and an Evaluation and Action Module (EAM). The CCM is triggered periodically (e.g. every 1:00 AM) to search for newly acquired CBCTs of patients of interest and then export the DICOM files of the images and related registrations defined in ARIA followed by triggering the PRM. The PRM imports the DICOM images and registrations, links the CBCTs to the related treatment plan of the patient in the planning system (RayStation V4.5, RaySearch, Stockholm, Sweden). A pre-determined CT-to-density table is automatically generated for dose calculation. Current version of the DCM uses a rigid registration which regards the treatment isocenter of the CBCT to be the isocenter of the treatment plan. Then it starts the dose calculation automatically. The AEM evaluates the plan using pre-determined plan evaluation parameters: PTV dose-volume metrics and critical organ doses. The tool has been tested for 10 patients. Results: Automatic plans are generated and saved in the order of the treatment dates of the Adaptive Planning module of the RayStation planning system, without any manual intervention. Once the CTV dose deviates more than 3%, both email and page alerts are sent to the physician and the physicist of the patient so that one can look the case closely. Conclusion: The tool is capable to perform automatic dose tracking and to alert clinicians when an action is needed. It is clinically useful for off-line adaptive therapy to catch any gross error. Practical way of determining alarming level for OAR is under development.

Purpose: To test a novel total body irradiation (TBI) system using conformal partial arc with patient lying on the stationary couch which is biologically equivalent to a moving couch TBI. This improves the scanning field TBI, which is previously presented. Methods: The Uniform MU Modulated arc Segments TBI or UMMS-TBI scans the treatment plane with a constant machine dose rate and a constant gantry rotation speed. A dynamic MLC pattern which moves while gantry rotates has been designed so that the treatment field moves same distance at the treatment plane per each gantry angle, while maintaining same treatment field size (34cm) at the plane. Dose across the plane varies due to the geometric differences including the distance from the source to a point of interest and the different attenuation from the slanted depth which changes the effective depth. Beam intensity is modulated to correct the dose variation across the plane by assigning the number of gantry angles inversely proportional to the uncorrected dose. Results: Measured dose and calculated dose matched within 1 % for central axis and 3% for off axis for various patient scenarios. Dose from different distance does not follow the inverse square relation as it is predicted from calculation. Dose uniformity better than 5% across 180 cm at 10cm depth is achieved by moving the gantry from −55 to +55 deg. Total treatment time for 2 Gy AP/PA fields is 40–50 minutes excluding patient set up time, at the machine dose rate of 200 MU/min. Conclusion: This novel technique, yet accurate but easy to implement enables TBI treatment in a small treatment room with less program development preparation than other techniques. The VMAT function of treatment delivery is not required to modulate beams. One delivery pattern can be used for different patients by changing the monitor units.

Purpose: Building up a TG-71 based electron monitor-unit (MU) calculation protocol usually involves massive measurements. This work investigates a minimum data set of measurements and its calculation accuracy and measurement time. Methods: For 6, 9, 12, 16, and 20 MeV of our Varian Clinac-Series linear accelerators, the complete measurements were performed at different depth using 5 square applicators (6, 10, 15, 20 and 25 cm) with different cutouts (2, 3, 4, 6, 10, 15 and 20 cm up to applicator size) for 5 different SSD's. For each energy, there were 8 PDD scans and 150 point measurements for applicator factors, cutout factors and effective SSDs that were then converted to air-gap factors for SSD 99–110cm. The dependence of each dosimetric quantity on field size and SSD was examined to determine the minimum data set of measurements as a subset of the complete measurements. The “missing” data excluded in the minimum data set were approximated by linear or polynomial fitting functions based on the included data. The total measurement time and the calculated electron MU using the minimum and the complete data sets were compared. Results: The minimum data set includes 4 or 5 PDD's and 51 to 66 point measurements for each electron energy, and more PDD's and fewer point measurements are generally needed as energy increases. Using only

Purpose: To reconstruct on-treatment 4D CT images from planning 4D CT images by adapting deformation vector field (DVF) of the planning CT to the on-treatment condition, while the adaptation is based on the scaling of two amplitudes that are motion characteristics at the times of treatment and planning CT acquisition, respectively. Methods: An anthropomorphic digital phantom (XCAT) was used to generate 4D image sets with 1-cm and 2-cm tumor motions simulating conditions of planning CT and treatment, respectively. DVFs were acquired from the planning CT image set. The DRR images were acquired simulating setup kV images from the two CT image sets. On the DRR images, tumor positions and their motion amplitudes were quantified. The DVFs were scaled linearly by the amplitude ratio between the treatment and the planning CT times, assuming the elasticity of lung. The scaled DVFs were used to resample the planning 4D CT images generating on-treatment 4D CT images. The on-treatment 4D CT images thus acquired were compared with the reference on-treatment images (2-cm motion). Results: The resampled images showed good agreement within 1 mm residual errors with the reference images. The normalized cross correlation was 0.995. Conclusion: A linear model of amplitude scaling was developed to reconstruct on-treatment 4D CT images from planning 4D CT images using the setup KV images acquired during treatment. The model was validated on a digital phantom. For the model to fully work, a further research needs to be followed, that aims at utilizing a phase-specific CT image set that is geometrically identical between pretreatment and treatment conditions.

Purpose: Appropriate MLC QA is important to assure treatment accuracy. In addition to the TG-142 recommended picket fence test, our institution performs four MLC tests on a weekly basis. The purpose of this study was to evaluate the clinical necessity of our weekly MLC QA program after a two-year run. Methods: Our Weekly MLC QA tests: 1) MLC positioning; 2) Interdigitation for testing hard MLC motion; 3) Picket fence at static gantry angle; 4) Leaf-end leakage and 5) dose match between a VMAT plan and the same plan with static-gantry-sliding window. For this study, a total of 794 Weekly MLC QA plans using EPID systems (3878 in total) for 10 machines from June 2014 to February 2016 were analyzed. Failure Mode Effect Analysis (FMEA) was performed with five failure modes. For each failure mode, a risk probability number (RPN) was calculated for a Weekly MLC QA (Monthly + Weekly) and for a Monthly MLC QA Only scheme. The probability of occurrence was evaluated from the Weekly QA experience. Results: Total number of failures for 794 cases were 24 (0.23%, for MLC Position test, 0.11% for Interdigitation test, 0.11% for Picket Fence Test, 0% for Leaf end Leakage test and 2.5% for VMAT test). The QA goal was to maintain the RPN score < 125. The RPN scores for the Monthly MLC QA Only ranged from 144–180, while those with Weekly (Monthly + Weekly) MLC QA decreased to 96–120. Conclusion: The Weekly MLC QA decreased the RPN numbers to the desired level (RPN

Purpose: This study compares the geometric-based setup (GBS) which is currently used in the clinic to a novel concept of dose-based setup (DBS) of head and neck (HN and evaluates the clinical advantages. Methods: Ten H&N patients who underwent re-simulation and re-plan due to noticeable anatomic changes during the course of the treatments were retrospectively reviewed on dosimetric changes in the assumption of no plan modification was performed. RayStation planning system (RaySearch Laboratories AB, Sweden) was used to match (ROI fusion module) between prescribed isodoseline (IDL) in the CBCT imported along with ROIs from re-planned CT and the IDL of original plan (Dose-based setup: DBS). Then, the CBCT plan based on daily setup using the GBS (previously used for a patient) and the DBS CBCT plan recalculated in RayStation compared against the original CT-sim plan. Results: Most of patients’ tumor coverage and OAR doses got generally worsen when the CBCT plans were compared with original CT-sim plan with GBS. However, when DBS intervened, the OAR dose and tumor coverage was better than the GBS. For example, one of patients’ daily average doses of right parotid and oral cavity increased to 26% and 36%, respectively from the original plan to the GBS planning. However, it only increased by 13% and 24%, respectively with DBS. GTV D95 coverage also decreased by 16% with GBS, but only 2% decreased with DBS. Conclusion: DBS method is superior to GBS to prevent any abrupt dose changes to OARs as well as PTV/CTV or GTV at least for some H&N cases. Since it is not known when the DBS is beneficial to the GBS, a system which enables the on-line DBS may be helpful for better treatment of H&N.

Purpose: To implement a comprehensive non-measurement-based verification program for patient-specific IMRT QA Methods: Based on published guidelines, a robust IMRT QA program should assess the following components: 1) accuracy of dose calculation, 2) accuracy of data transfer from the treatment planning system (TPS) to the record-and-verify (RV) system, 3) treatment plan deliverability, and 4) accuracy of plan delivery. Results: We have implemented an IMRT QA program that consist of four components: 1) an independent re-calculation of the dose distribution in the patient anatomy with a commercial secondary dose calculation program: Mobius3D (Mobius Medical Systems, Houston, TX), with dose accuracy evaluation using gamma analysis, PTV mean dose, PTV coverage to 95%, and organ-at-risk mean dose; 2) an automated in-house-developed plan comparison system that compares all relevant plan parameters such as MU, MLC position, beam iso-center position, collimator, gantry, couch, field size settings, and bolus placement, etc. between the plan and the RV system; 3) use of the RV system to check the plan deliverability and further confirm using “mode-up” function on treatment console for plans receiving warning; and 4) implementation of a comprehensive weekly MLC QA, in addition to routine accelerator monthly and daily QA. Among 1200 verifications, there were 9 cases of suspicious calculations, 5 cases of delivery failure, no data transfer errors, and no failure of weekly MLC QA. These 9 suspicious cases were due to the PTV extending to the skin or to heterogeneity correction effects, which would not have been caught using phantom measurement-based QA. The delivery failure was due to the rounding variation of MLC position between the planning system and RV system. Conclusion: A very efficient, yet comprehensive, non-measurement-based patient-specific QA program has been implemented and used clinically for about 18 months with excellent results

Purpose: To report practical issues and solutions in reconstructing and using 4DCT to account for respiratory motion in radiotherapy planning. Methods: Quiet breathing 4DCT was used to account for respiratory motion for patients with lung or upper abdomen tumor. A planning CT and a 4DCT were acquired consecutively with a Philips Brilliance CT scanner and Varian RPM System. The projections were reconstructed into 10 phases. In Pinnacle RTP system, we contour a GTV in each phase and unite all 10 GTVs as ITV. The ITV is then mapped to the planning CT. We describe practical issues, their causes, our solutions and reasoning during this process. Results: In 6 months, 9 issues were reported for 8 patients with lung cancer. For two patients, part of the GTV (∼50% and 10%) in planning CT fell outside the ITV in 4DCT. There was a 7 mm variation in first patient back position because less restricted immobilization had to be used. The second discrepancy was due to moderate variation in breathing amplitude. We extended the ITV to include the GTV since both variations may likely happen during treatment. A LUL tumor showed no motion due to a 10-s long no-breathing period. An RLL tumor appeared double due to an abnormally deeper breath at the tumor region. We repeated 4DCT reiterating the importance of quiet, regular breathing. One patient breathed too light to generate RPM signal. Two issues (no motion in lung, incomplete images in 90% phase) were due to incorrect tag positions. Two unexplainable errors disappeared when repeating reconstruction. In summary, 5 issues were patient-related and 4 were technique issues. Conclusion: Improving breathing regularity avoided large artifacts in 4DCT. One needs to closely monitor patient breathing. For uncontrollable variations, larger PTVs are necessary which requires appropriate communication between physics and the treating physician.

Purpose: Enable a scanning field total body irradiation (TBI) technique, using dynamic arcs, which is biologically equivalent to a moving couch TBI. Methods: Patient is treated slightly above the floor and the treatment field scans across the patient by a moving gantry. MLC positions change during gantry motion to keep same field opening at the level of the treatment plane (170 cm). This is done to mimic the same geometry as the moving couch TBI technique which has been used in our institution for over 10 years. The dose rate and the gantry speed are determined considering a constant speed of the moving field, variations in SSD and slanted depths resulting from oblique gantry angles. An Eclipse (Varian) planning system is commissioned to accommodate the extended SSD. The dosimetric foundations of the technique have been thoroughly investigated using phantom measurements. Results: Dose uniformity better than 2% across 180 cm length at 10cm depth is achieved by moving the gantry from −55 to +55 deg. Treatment range can be extended by increasing gantry range. No device such as a gravity-oriented compensator is needed to achieve a uniform dose. It is feasible to modify the dose distribution by adjusting the dose rate at each gantry angle to compensate for body thickness differences. Total treatment time for 2 Gy AP/PA fields is 40–50 minutes excluding patient set up time, at the machine dose rate of 100 MU/min. Conclusion: This novel yet transportable moving field technique enables TBI treatment in a small treatment room with less program development preparation than other techniques. Treatment length can be extended per need, and. MLC-based thickness compensation and partial lung blocking are also possible.

Purpose: to mitigate artifacts induced by the presence of the RT treatment couch in on-board CBCT and improve image quality Methods: a model of a Varian IGRT couch is constructed using a CBCT scan of the couch in air. The model is used to generate a set of forward projections (FP) of the treatment couch at specified gantry angles. The model couch forward projections are then used to process CBCT scan projections which contain the couch in addition to the scan object (Catphan phantom), in order to remove the attenuation component of the couch at any given gantry angle. Prior to pre-processing with the model FP, the Catphan projection data is normalized to an air scan with bowtie filter. The filtered Catphan projections are used to reconstruct the CBCT with an in-house FDK algorithm. The artifact reduction in the processed CBCT scan is assessed visually, and the image quality improvement is measured with the CNR over a few selected ROIs of the Catphan modules. Results: Sufficient match between the forward projected data and the x-ray projections is achieved to allow filtering in attenuation space. Visual improvement of the couch induced artifacts is achieved, with a moderate expense of CNR. Conclusion:more » Couch model-based correction of CBCT projection data has a potential for qualitative improvement of clinical CBCT scans, without requiring position specific correction data. The technique could be used to produce models of other artifact inducing devices, such as immobilization boards, and reduce their impact on patient CBCT images.« less

Purpose: To develop and to test a method to generate a new 4D CT images of the treatment day from the old 4D CT and the portal images of the day when the motion extent exceeded from that represented by plan CTs. Methods: A motion vector of a moving tumor in a patient may be extended to reconstruct the tumor position when the motion extent exceeded from that represented by plan CTs. To test this, 1. a phantom that consists of a polystyrene cylinder (tumor) embedded in cork (lung) was placed on a moving platform with 4 sec/cycle and amplitudes of 1 cm and 2 cm, and was 4D-scanned. 2. A 6MV photon beam was irradiated on the moving phantoms and cineEPID images were obtained. 3. A motion vector of the tumor was acquired from 4D CT images of the phantom with 1 cm amplitude. 4. From cine EPID images of the phantom with the 2 cm amplitude, various motion extents (0.3 cm, 0.5 cm, etc) were acquired and programmed into the motion vector, producing CT images at each position. 5. The reconstructed CT images were then compared with pre-acquired “reference” 4D CT images at each position (i.e. phase). Results: The CT image was reconstructed and compared with the reference image, showing a slight mismatch in the transition direction limited by voxel size (slice thickness) in CT image. Due to the rigid nature of the phantom studied, the modeling the displacement of the center of object was sufficient. When deformable tumors are to be modeled, more complex scheme is necessary, which utilize cine EPID and 4D CT images. Conclusion: The new idea of CT image reconstruction was demonstrated. Deformable tumor movements need to be considered in the future.

Purpose: A dedicated stereotactic breast radiotherapy device, GammaPod, was developed to treat early stage breast cancer. The first clinical GammaPod unit is currently being commissioned to treat its first patient in summer 2015. We provide an absorbed dose calibration formalism following AAPM protocols with modifications for the calibration medium and unique geometry of GammaPod. We also provide cross-comparison of the dosimetric calibration using different types of dosimeters. Methods: GammaPod is comprised of a rotating hemi-spherical source carrier containing 36 Co-60 sources and a concentric rotating tungsten collimator providing beam diameters of 1.5 and 2.5 cm. A solid breast phantom made of ultra-high-molecular-weight polyethylene (density 0.935 g/cm3) with an effective depth of 7cm is used for absolute dose calibration. Due to the non-water equivalent medium, we utilized the former AAPM TG-21 protocol, but cross-compared the calibration with TG-51 calibrated OSLD and radiochromic film dosimeters. Polyethylene medium and chamber-specific calibration parameters are derived and provided for Exradin A-18 and Capintec PR-05P ionization chambers. Results: The first clinical unit was installed with a total Co-60 activity of 2672 Ci in November 2014. Maximum dose rate was determined to be 2.76 Gy/min and 2.63 Gy/min for 2.5 and 1.5 cm collimators respectively at 7 cm depth. The determination of Ngas and appropriate polyethylene dosimetric quantities (stopping power and absorption coefficient ratios) is performed using NIST data tables and supplied as reference for calibration of subsequent units. Calibration crosscheck was accomplished with OSLD and radiochromic films, both with secondary TG-51 based calibrations, yielding relative dose differences less than 1%. Conclusion: The first GammaPod treatment unit for breast stereotactic radiotherapy was successfully installed and calibrated for commencement of patient treatments in 2015. An absolute dose calibration formalism was established for the unique geometry of GammaPod and the method is validated with different detectors. P Hoban, Y Niu and C Yu are employees of Xcision Medical Systems.

Purpose: To assess dose calculation accuracy of cone-beam CT (CBCT) based treatment plans using a patient-specific stepwise CT-density conversion table in comparison to conventional CT-based treatment plans. Methods: Unlike CT-based treatment planning which use fixed CT-density table, this study used patient-specific CT-density table to minimize the errors in reconstructed mass densities due to the effects of CBCT Hounsfield unit (HU) uncertainties. The patient-specific CT-density table was a stepwise function which maps HUs to only 6 classes of materials with different mass densities: air (0.00121g/cm3), lung (0.26g/cm3), adipose (0.95g/cm3), tissue (1.05 g/cm3), cartilage/bone (1.6g/cm3), and other (3g/cm3). HU thresholds to define different materials were adjusted for each CBCT via best match with the known tissue types in these images. Dose distributions were compared between CT-based plans and CBCT-based plans (IMRT/VMAT) for four types of treatment sites: head and neck (HN), lung, pancreas, and pelvis. For dosimetric comparison, PTV mean dose in both plans were compared. A gamma analysis was also performed to directly compare dosimetry in the two plans. Results: Compared to CT-based plans, the differences for PTV mean dose were 0.1% for pelvis, 1.1% for pancreas, 1.8% for lung, and −2.5% for HN in CBCT-based plans. The gamma passing ratemore » was 99.8% for pelvis, 99.6% for pancreas, and 99.3% for lung with 3%/3mm criteria, and 80.5% for head and neck with 5%/3mm criteria. Different dosimetry accuracy level was observed: 1% for pelvis, 3% for lung and pancreas, and 5% for head and neck. Conclusion: By converting CBCT data to 6 classes of materials for dose calculation, 3% of dose calculation accuracy can be achieved for anatomical sites studied here, except HN which had a 5% accuracy. CBCT-based treatment planning using a patient-specific stepwise CT-density table can facilitate the evaluation of dosimetry changes resulting from variation in patient anatomy.« less

Purpose: Raven QA (JPLC, MD) is a unified and comprehensive quality assurance system for QA of TG-142, which use a phosphor screen, a mirror system and a camera. It is to test if this device can be used for IMRT QA dosimetry. Methods: A lung IMRT case is used deliver dose to Raven QA. Accuracy of dose distribution of 5cm slab phantom using Eclipse planning system (Varian) has been confirmed both from a Monte Carlo Simulation and from a MapCheck (SunNuclear) measurement. Geometric distortion and variation of spatial dose response are corrected after background subtraction. A pin-hole grid plate is designed and used to determine the light scatter in the Raven QA box and the spatial dose response. Optic scatter model was not applied in this preliminary study. Dose is normalized to the response of the 10×10 field and the TMR of 5cm depth was considered. Results: Time to setup the device for IMRT QA takes less than 5 minutes as other commercially available devices. It shows excellent dose linearity and dose rate independent, less than 1 %. Background signal, however, changes for different field sizes. It is believed to be due to inaccurate correction of optic scatter. Absolute gammamore » (5%, 5mm) passing rate was higher than 95%. Conclusion: This study proves that the Raven QA can be used for a patient specific IMRT verification. Part of this study is supported by the Maryland Industrial Partnership Grant.« less

Purpose: This work is to investigate the feasibility of a standardized monthly quality check (QC) of LINAC output determination in a multi-site, multi-LINAC institution. The QC was developed to determine individual LINAC output using the same optimized measurement setup and a constant calibration factor for all machines across the institution. Methods: The QA data over 4 years of 7 Varian machines over four sites, were analyzed. The monthly output constancy checks were performed using a fixed source-to-chamber-distance (SCD), with no couch position adjustment throughout the measurement cycle for all the photon energies: 6 and 18MV, and electron energies: 6, 9, 12, 16 and 20 MeV. The constant monthly output calibration factor (Nconst) was determined by averaging the machines’ output data, acquired with the same monthly ion chamber. If a different monthly ion chamber was used, Nconst was then re-normalized to consider its different NDW,Co-60. Here, the possible changes of Nconst over 4 years have been tracked, and the precision of output results based on this standardized monthly QA program relative to the TG-51 calibration for each machine was calculated. Any outlier of the group was investigated. Results: The possible changes of Nconst varied between 0–0.9% over 4 years. The normalization of absorbed-dose-to-water calibration factors corrects for up to 3.3% variations of different monthly QA chambers. The LINAC output precision based on this standardized monthly QC relative to the TG-51 output calibration is within 1% for 6MV photon energy and 2% for 18MV and all the electron energies. A human error in one TG-51 report was found through a close scrutiny of outlier data. Conclusion: This standardized QC allows for a reasonably simplified, precise and robust monthly LINAC output constancy check, with the increased sensitivity needed to detect possible human errors and machine problems.

Purpose: Only a part of a treatment couch is reconstructed in CBCT due to the limited field of view (FOV). This often generates inaccurate results in the delivered dose evaluation with CBCT and more noise in the CBCT reconstruction. Full reconstruction of the couch at treatment setup can be used for more accurate exit beam dosimetry. The goal of this study is to develop a method to reconstruct a full treatment couch using a pre-scanned couch image and rigid registration. Methods: A full couch (Exact Couch, Varian) model image was reconstructed by rigidly registering and combining two sets of partial CBCT images. The full couch model includes three parts: two side rails and a couch top. A patient CBCT was reconstructed with reconstruction grid size larger than the physical field of view to include the full couch. The image quality of the couch is not good due to data truncation, but good enough to allow rigid registration of the couch. A composite CBCT image of the patient plus couch has been generated from the original reconstruction by replacing couch portion with the pre-acquired model couch, rigidly registered to the original scan. We evaluated the clinical usefulness of this method bymore » comparing treatment plans generated on the original and on the modified scans. Results: The full couch model could be attached to a patient CBCT image set via rigid image registration. Plan DVHs showed 1∼2% difference between plans with and without full couch modeling. Conclusion: The proposed method generated a full treatment couch CBCT model, which can be successfully registered to the original patient image. This method was also shown to be useful in generating more accurate dose distributions, by lowering 1∼2% dose in PTV and a few other critical organs. Part of this study is supported by NIH R01CA133539.« less

Purpose: To develop a CBCT HU correction method using a patient specific HU to mass density conversion curve based on a novel image registration and organ mapping method for head-and-neck radiation therapy. Methods: There are three steps to generate a patient specific CBCT HU to mass density conversion curve. First, we developed a novel robust image registration method based on sparseness analysis to register the planning CT (PCT) and the CBCT. Second, a novel organ mapping method was developed to transfer the organs at risk (OAR) contours from the PCT to the CBCT and corresponding mean HU values of each OAR were measured in both the PCT and CBCT volumes. Third, a set of PCT and CBCT HU to mass density conversion curves were created based on the mean HU values of OARs and the corresponding mass density of the OAR in the PCT. Then, we compared our proposed conversion curve with the traditional Catphan phantom based CBCT HU to mass density calibration curve. Both curves were input into the treatment planning system (TPS) for dose calculation. Last, the PTV and OAR doses, DVH and dose distributions of CBCT plans are compared to the original treatment plan. Results: One head-and-neckmore » cases which contained a pair of PCT and CBCT was used. The dose differences between the PCT and CBCT plans using the proposed method are −1.33% for the mean PTV, 0.06% for PTV D95%, and −0.56% for the left neck. The dose differences between plans of PCT and CBCT corrected using the CATPhan based method are −4.39% for mean PTV, 4.07% for PTV D95%, and −2.01% for the left neck. Conclusion: The proposed CBCT HU correction method achieves better agreement with the original treatment plan compared to the traditional CATPhan based calibration method.« less

Purpose: To: (1) describe an independent, automated, systematic software-based protocol for verifying clinical data accuracy/integrity for mitigation of data corruption/loss risks following radiation oncology information system (ROIS) upgrades; and (2) report on application of this approach in an academic/community practice environment. Methods: We propose a robust approach to perform quality assurance on the ROIS after an upgrade, targeting four data sources: (1) ROIS relational database; (2) ROIS DICOM interface; (3) ROIS treatment machine data configuration; and (4) ROIS-generated clinical reports. We investigated the database schema for differences between pre-/post-upgrade states. Paired DICOM data streams for the same object (such as RT-Plan/Treatment Record) were compared between pre-/post-upgrade states for data corruption. We examined machine configuration and related commissioning data files for changes and corruption. ROIS-generated treatment appointment and treatment parameter reports were compared to ensure patient encounter and treatment plan accuracy. This protocol was supplemented by an end-to-end clinical workflow test to verify essential ROI functionality and integrity of components interfaced during patient care chain of activities. We describe the implementation of this protocol during a Varian ARIA system upgrade at our clinic. Results: We verified 1,638 data tables with 2.4 billion data records. For 222 under-treatment patients, 605 DICOM RT plans and 13,480 DICOM treatment records retrieved from the ROIS DICOM interface were compared, with no differences in fractions, doses delivered, or treatment parameters. We identified 82 new data tables and 78 amended/deleted tables consistent with the upgrade. Reports for 5,073 patient encounters over a 2-week horizon were compared and were identical to those before the upgrade. Content in 12,237 xml machine files was compared, with no differences identified. Conclusion: An independent QA/validation approach for ROIS upgrades was developed and implemented at our clinic. The success of this approach ensures a robust QA of ROIS upgrades without manual paper/electronic checks and associated intensive labor.

Purpose: It is important to estimate dose to pacemaker (PM)/Implantable Cardioverter Defibrillator (ICD) before undertaking Accelerated Partial Breast Treatment using High Dose Rate (HDR) brachytherapy. Kim et al. have reported HDR PM/ICD dose using a single-source balloon applicator. To the authors knowledge, there have so far not been any published PM/ICD dosimetry literature for the Strut Adjusted Volume Implant (SAVI, Cianna Medical, Aliso Viejo, CA). This study aims to fill this gap by generating a dose look up table (LUT) to predict maximum dose to the PM/ICD in SAVI HDR brachytherapy. Methods: CT scans for 3D dosimetric planning were acquired for four SAVI applicators (6−1-mini, 6−1, 8−1 and 10−1) expanded to their maximum diameter in air. The CT datasets were imported into the Elekta Oncentra TPS for planning and each applicator was digitized in a multiplanar reconstruction window. A dose of 340 cGy was prescribed to the surface of a 1 cm expansion of the SAVI applicator cavity. Cartesian coordinates of the digitized applicator were determined in the treatment leading to the generation of a dose distribution and corresponding distance-dose prediction look up table (LUT) for distances from 2 to 15 cm (6-mini) and 2 to 20 cm (10–1).The deviation between the LUT doses and the dose to the cardiac device in a clinical case was evaluated. Results: Distance-dose look up table were compared to clinical SAVI plan and the discrepancy between the max dose predicted by the LUT and the clinical plan was found to be in the range (−0.44%, 0.74%) of the prescription dose. Conclusion: The distance-dose look up tables for SAVI applicators can be used to estimate the maximum dose to the ICD/PM, with a potential usefulness for quick assessment of dose to the cardiac device prior to applicator placement.

Purpose: To test a method to reconstruct a four-dimensional (4D) dose distribution using the correlation of pre-calculated 4D electronic portal imaging device (EPID) images and measured cine-EPID images. Methods: 1. A phantom designed to simulate a tumor in lung (a polystyrene block with 3.0 cm diameter embedded in cork) was placed on a sinusoidally moving platform with 2 cm amplitude and 4 sec/cycle. Ten-phase 4D CT images were acquired for treatment planning and dose reconstruction. A 6MV photon beam was irradiated on the phantom with static (field size=5×8.5 cm2) and dynamic fields (sliding windows, 10×10 cm2, X1 MLC closing in parallel with the tumor movement). 2. 4D and 3D doses were calculated forwardly on PTV (1 cm margin). 3. Dose images on EPID under the fields were calculated for 10 phases. 4. Cine EPID images were acquired during irradiation. 5. Their acquisition times were correlated to the phases of the phantom at which irradiation occurred by inter-comparing calculated “reference” EPID images with measured images (2D gamma comparison). For the dynamic beam, the tumor was hidden under MLCs during a portion of irradiation time; the correlation performed when the tumor was visible was extrapolated. 6. Dose for each phase was reconstructed on the 4D CT images and summed over all phases. The summation was compared with forwardly calculated 4D and 3D dose distributions. Monte Carlo methods were used for all calculations. Results: For the open and dynamic beams, the 4D reconstructed doses showed the pass rates of 92.7 % and 100 %, respectively, at the isocenter plane given 3% / 3 mm criteria. The better agreement of the dynamic beam was from its dose gradient which blurred the otherwise sharp difference between forward and reconstructed doses. This also contributed slightly better agreement in DVH of PTV. Conclusion: The feasibility of 4D reconstruction was demonstrated.

Purpose: Dose rate regulated Tracking (DRRT) is known to be an efficient method which adaptively deliver tracking treatments when patient breathing is irregular. The interface for the dose rate controlling, however, is not commercially available. The Motion Management Interface (MMI, Varian Medical System, CA), which provides beam on/off switching during treatment is available for clinic. This study is to test if delivering the adaptive tumor tracking is feasible for irregular breathing using beam switching with MMI. Methods: Fifty‐five free breathing RPM traces acquired from lung cancer patients are used. The first day RPM traces of the patients are utilized to design preprogrammed tracking MLC patterns, of which periods are intentionally reduced by 20% in order to catch up the variation of patient breathing irregularity in the treatment day. Eligibility criteria for this technique is variation of amplitude and period less than 20%. An algorithm which determines beam on/off every 100 ms by considering the preprogrammed (MLC) positions and current breathing positions is developed. Tracking error and delivery efficiency is calculated by simulating the beam‐switching adaptive tracking from the RPM traces. Results: Breathing patterns of 38 patients (70%) met the eligibility criteria. Tracking errors of all of the cases who meet the criteria are less than 2mm (average 1.4mm) and the average delivery efficiency was 71%. Those of rest of the cases are 1.9 mm and 48%. Conclusions: Beam switching is effectively equivalent to the dose rate regulation for DRRT. Adaptive tracking with beam switching is feasible if patient selection is based on the eligibility criteria. NIH R01CA133539

Respiration causes target motion, which is known to be one of the technical bottlenecks in radiotherapy, especially for stereotactic radio-surgery and intensity modulated radiotherapy (IMRT). To overcome this problem, aperture maneuver with compelled breath (AMC) has been developed. In order to simulate compelled respiratory motion, a moving phantom using a ventilator was designed. As the air flow was forced to the bellows, which simulates the lungs, by a ventilator, a film connected to the ventilator moved like the respiratory target motion. A software was developed to transfer multileaf collimator motion from breathless to actual periodic breathing conditions. Static fields as well as step-and-shoot IMRT fields were modified in accordance with moving shapes to follow the target position, using the software with the controlled breathing information. Film dosimetry for a small field and for IMRT fields with a moving phantom was performed. To evaluate clinical implementation, five healthy volunteers were tested to breathe through a ventilator, and all of them could adapt the compelled breath without any difficulties. Additive margins for a moving target with AMC were not larger than 3 mm for respiratory organ motions up to 18 mm, while those with the static beam were 9 mm. For IMRT fields, large discrepancies were present between a static target and a moving target with the static beam, while they coincided well with AMC. Clinical acceptable differences between the dose distributions from a static target with the static beam and from a moving target with AMC revealed that this technique could be applied clinically.

Purpose: We propose a practical four‐dimensional (4D) planning scheme of intensity‐modulated arc therapy (IMAT) for tumor tracking using dynamic multileaf collimators(MLC). Methods: We contour the target and critical structures on each image set of 4DCT associated with each breathing phase. Transformation vectors of the target from a reference phase to other phases are derived. A 3D‐IMAT plan is first optimized on the reference phase without motion‐induced margin. Assuming delivery starts at a predetermined breathing phase and the patient breathes the same way as during 4DCT imaging, we know the phases at which each planned segment will be delivered. Then, we transform the optimized segments from the reference phase to the phases of their delivery based on projections of transformation vectors of the target in the beam's eye‐view. The connectivity of the transformed segments between adjacent beam angles is enforced. The dose for each phase is calculated using segments delivered at that phase, and transformed to the reference phase to obtain cumulative dose. The resulting 4D plan is compared with the 3D plan without considering respiratory motion and with motion‐induced margins. Effects of enforcing connectivity on the plan quality are also evaluated. Results: Compared to the PTV dose of 3D plan without tumor tracking, the 4D PTV dose is more conformal and uniform, and it approaches the ideal situation of the patient being static. The 4D plan also showed better critical organ sparing than the 3D plan with motion‐induced margins. Adding MLC motion required for tracking to the planned motion rarely causes physical constraints to be exceeded, and enforcing connectivity thus has little effect on plan quality. Conclusions: The proposed scheme of 4D‐IMAT planning for tumor tracking is a feasible and practical approach under the current clinical planning environment. Dosimetric benefits of 4D planning have been clearly demonstrated. This stusy is supported in part by NIH R01 grant 1R01CA133539‐01A2

Purpose: To simulate patient flow in proton treatment center under uncertainty and to explore the feasibility of treatment preparation rooms to improve patient throughput and cyclotron utilization. Methods: Three center layout scenarios were modeled: (S1: In-Tx room imaging) patient setup and imaging (planar/volumetric) performed in treatment room, (S2: Patient setup in preparation room) each treatment room was assigned with preparation room(s) that was equipped with lasers only for patient setup and gross patient alignment, and (S3: Patient setup and imaging in preparation room) preparation room(s) was equipped with laser and volumetric imaging for patient setup, gross and fine patient alignment. A 'snap' imaging was performed in treatment room. For each scenario, the number of treatment rooms and the number of preparation rooms serving each treatment room were varied. We examined our results (average of 100 16-hour (two shifts) working days) by evaluating patient throughput and cyclotron utilization. Results: When the number of treatment rooms increased ([from, to]) [1, 5], daily patient throughput increased [32, 161], [29, 184] and [27, 184] and cyclotron utilization increased [13%, 85%], [12%, 98%], and [11%, 98%] for scenarios S1, S2 and S3 respectively. However, both measures plateaued after 4 rooms. With the preparation rooms, themore » throughput and the cyclotron utilization increased by 14% and 15%, respectively. Three preparation rooms were optimal to serve 1-3 treatment rooms and two preparation rooms were optimal to serve 4 or 5 treatment rooms. Conclusion: Patient preparation rooms for patient setup may increase throughput and decrease the need for additional treatment rooms (cost effective). Optimal number of preparation rooms serving each gantry room varies as a function of treatment rooms and patient setup scenarios. A 5th treatment room may not be justified by throughput or utilization.« less

Purpose: Increasing numbers of institutions are beginning to utilize calculation-based methods (Monte Carlo or commercial software) of performing patient specific IMRT QA. In such systems the full three-dimensional (3-D) dose distribution may be available. In this work we determine the gamma pass rate calculated using the full 3-D dose distribution that is equivalent to the more conventional 2-D gamma pass rate. Methods: 55 IMRT and SBRT plans with a variety of field sizes were selected. All plans were created using Pinnacle v9.0. The delivered dose for each plan was re-computed using Monte Carlo simulation and then compared to the planned dose. 2-D gamma analysis was performed on each of three planes (axial, sagittal and coronal) and the values were compared with those obtained through 3-D gamma analysis. Gamma criteria used included 2%/2mm, 3%/3mm and 5%/5mm. Results: 3-D and 2-D gamma were linearly related (R2 = 0.84). The linear relationship held for all three 2-D planes and did not appear to depend on field size or on gamma criterion. For a given 2-D gamma value, the calculated 3-D gamma was always larger. For a 2-D gamma pass rate of 95% (range 92–97%), the average equivalent 3-D gamma was 97.5% (range 94.3 – 99.7%). Conclusion: For patient-specific IMRT QA involving the full 3-D dose distribution, acceptable plans require a larger percentage of points to pass gamma analysis than for the equivalent 2-D analysis employed during measurement-based QA. If a plan shows a 95% pass rate for a 2-D planar gamma calculation, the corresponding acceptable 3-D plan will have a pass rate of 97.5%. Clinics that use the full 3-D dose distribution for QA will need to adjust their passing criteria accordingly.

Purpose: The Iterative Subtraction Algorithm (ISA) method generates retrospectively a pre-selected motion phase cone-beam CT image from the full motion cone-beam CT acquired at standard rotation speed. This work evaluates ISA method with real lung patient data. Methods: The goal of the ISA algorithm is to extract motion and no- motion components form the full reconstruction CBCT. The workflow consists of subtracting from the full CBCT all of the undesired motion phases and obtain a motion de-blurred single-phase CBCT image, followed by iteration of this subtraction process. ISA is realized as follows: 1) The projections are sorted to various phases, and from all phases, a full reconstruction is performed to generate an image CTM. 2) Generate forward projections of CTM at the desired phase projection angles, the subtraction of projection and the forward projection will reconstruct a CTSub1, which diminishes the desired phase component. 3) By adding back the CTSub1 to CTm, no motion CBCT, CTS1, can be computed. 4) CTS1 still contains residual motion component. 5) This residual motion component can be further reduced by iteration.The ISA 4DCBCT technique was implemented using Varian Trilogy accelerator OBI system. To evaluate the method, a lung patient CBCT dataset was used. The reconstruction algorithm is FDK. Results: The single phase CBCT reconstruction generated via ISA successfully isolates the desired motion phase from the full motion CBCT, effectively reducing motion blur. It also shows improved image quality, with reduced streak artifacts with respect to the reconstructions from unprocessed phase-sorted projections only. Conclusion: A CBCT motion de-blurring algorithm, ISA, has been developed and evaluated with lung patient data. The algorithm allows improved visualization of a single phase motion extracted from a standard CBCT dataset. This study has been supported by National Institute of Health through R01CA133539.

Purpose: A method to generate a series of fluoro-tomographic images (FTI) of the slice of interest (SOI) from the projection images of the On-board imager (OBI) while gantry is moving is developed and tested. Methods: Tomographic image via background subtraction, TIBS has been published by our group. TIBS uses a priori anatomical information from a previous CT scan to isolate a SOI from a planar kV image by factoring out the attenuations by tissues outside the SOI (background). We extended the idea to 4D TIBS, which enables to generate from the projection of different gantry angles. A set of background images for different angles are prepared. A background image at a given gantry angle is subtracted from the projection image at the same angle to generate a TIBS image. Then the TIBS image is converted to a reference angle. The 4D TIBS is the set of TIBS that originated from gantry angles other than the reference angle. Projection images of lung patients for CBCT acquisition are used to test the 4D TIBS. Results: Fluoroscopic images of a coronal plane of lung patients are acquired from the CBCT projections at different gantry angles and times. Change of morphology of hilar vessels due to breathing and heart beating are visible in the coronal plane, which are generated from the set of the projection images at gantry angles other than antero-posterior. Breathing surrogate or sorting process is not needed. Unlike tomosynthesis, FTI from 4D TIBS maintains the independence of each of the projections thereby reveals temporal variations within the SOI. Conclusion: FTI, fluoroscopic imaging of a SOI with x-ray projections, directly generated from the x-ray projection images at different gantry angles is tested with a lung case and proven feasible. This technique can be used for on-line imaging of moving targets. NIH Grant R01CA133539

Purpose: To optimize the MLC model parameters for IMRT in the RayStation v.4.0 planning system and for a Varian C-series Linac with a 120-leaf Millennium MLC. Methods: The RayStation treatment planning system models rounded leaf-end MLC with the following parameters: average transmission, leaf-tip width, tongue-and-groove, and position offset. The position offset was provided by Varian. The leaf-tip width was iteratively evaluated by comparing computed and measured transverse dose profiles of MLC-defined fields at dmax in water. The profile comparison was also used to verify the MLC position offset. The transmission factor and leaf tongue width were derived iteratively by optimizing five clinical patient IMRT QA Results: brain, lung, pancreas, head-and-neck (HN), and prostate. The HN and prostate cases involved splitting fields. Verifications were performed with Mapcheck2 measurements and Monte Carlo calculations. Finally, the MLC model was validated using five test IMRT cases from the AAPM TG119 report. Absolute gamma analyses (3mm/3% and 2mm/2%) were applied. In addition, computed output factors for MLC-defined small fields (2×2, 3×3, 4×4, 6×6cm) of both 6MV and 18MV were compared to those measured by the Radiological Physics Center (RPC). Results: Both 6MV and 18MV models were determined to have the same MLC parameters: 2.5% transmission,more » tongue-and-groove 0.05cm, and leaftip 0.3cm. IMRT QA analysis for five cases in TG119 resulted in a 100% passing rate with 3mm/3% gamma analysis for 6MV, and >97.5% for 18MV. With 2mm/2% gamma analysis, the passing rate was >94.6% for 6MV and >90.9% for 18MV. The difference between computed output factors in RayStation and RPC measurements was less than 2% for all MLCdefined fields, which meets the RPC's acceptance criterion. Conclusion: The rounded leaf-end MLC model in RayStation 4.0 planning system was verified and IMRT commissioning was clinically acceptable. The IMRT commissioning was well validated using guidance from the AAPMTG119 protocol.« less

Purpose: This work aims (1) to determine the validity of the current practice of the phase-based 4DCT sorting (abdomen signal) for thorax scans; (2) to assess the potential benefit of utilizing the 3D surface breathing surrogate of chest area for phase-based 4DCT sorting. Methods: Dynamic 3D surface images over the chest and abdomen were acquired using 3D-suface-camera-system (VisionRT Ltd., London, UK) for a total of 10 patients. The vertical motions of five regions-of-interest (ROIs): abdomen, mid-sternum, lower-, middle-, and upper- ipsilateral-chest were tracked. Based on our previous work, the main source of motion artifact is the variation-of-phase-differences (VPD) between the breathing surrogate and the organ motion during scanning. Thus, we utilize VPD as a quantitative index of the variation of the resulting 4DCT images. The average VPD over a 60 second time-period was calculated for each of the abdomen and ROI combination. A simulation study with a wire phantom was designed to demonstrate the resulting 4DCT images. Results: Simulation showed significant image distortion or motion artifacts when VPD is larger than 5% of breathing period. The average VPDs of individual ROIs of each patient indicates that 75%, 59% and 22% cases have VPD larger than 5%, 10% and 15% of period each cycle between the selected ROI and the abdominal motions patterns, respectively (10.1±6.0 %, minimum 1.0% to maximum 23.9 %). Overall, patients with less 3-second breathing-period present VPD larger than 10%. The correlations between the abdomen and individual chest ROIs motions as well as the motions among chest ROIs are highly case-dependent. Conclusion: Abdominal motion does not represent the chest motions for some of the patients. This results in motion artifacts in the phase-based 4DCT. Utilizing the surface surrogate of the chest region can potentially improve phase-based 4DCT sorting for thorax 4DCT scan, especially for fast breathing patients.

Purpose: To show the feasibility of clinical application of DrMAT for SBRT in lung cancer patients. DrMAT is a form of dynamic conformal arc therapy where MLC segments and dose rates are controlled through simple field weight optimization. Methods: To show feasibility a new treatment plan was created based on the CT of SBRT lung cancer patients. Static plans with 33 fields are made, which have 11deg in between each field and are acquired rotating gantry angle from 180deg to 188deg in CCW direction, total 352deg is rotated. MLC maintained static aperture for each field. To optimize 33 individual fields, field weight was adjusted accordingly using weight optimization algorithm. Keeping weights and MU of static plan, static MLC aperture was converted to multiple arc segments. Arc plan could be created with the fields in the intervals of 11deg. Static MLC should be converted to arc segment MLC. Dynamic conformal arc therapy plan consists of 33 arc fields, is converted to one dose rate modulated arc therapy (DrMAT) plan. DrMAT plan consists of 166 control points which becomes a single arc plan that changes the shape of MLC for every 2.2deg. The resulting DrMAT plan is not an inverse plan it is a simple form of dynamic conformal arc plan using field weight obtained from static plan. This is compared and evaluated with the VMAT plan. Results: DrMAT and VMAT plans have been compared based on the RTOG1021. Both DrMAT and VMAT plans satisfy 100% irradiation to 95% of PTV and critical organs did not exceed dose limit suggested in RTOG1021. DrMAT plan is almost similar with VMAT plan in Result. Conclusion: Field weight optimization method did not show better Resultcompared to VMAT optimization. However, considering simplicity, DrMAT satisfies the condition in RTOG1021. Therefore clinical application of DrMAT is feasible.

Purpose: It has been always challenging to reconstruct good quality 4D-CT or 4D-CBCT when the data are under-sampled for each individual phase. For example, CBCT data is often acquired with limited number of projections and while it is sufficient to reconstruct decent motion averaged images; reconstructing a 4D image from this limited data leads to motion induced artifacts. However, inherent in the projection data are both components that capture the moving and non-moving parts of the image. In this work, we test the feasibility of using such a dataset to separate out the moving and non-moving components and then reconstruct individual phase images to acquire a complete 4D reconstruction with reduced artifacts. Methods: To test our theory, we simulated a simple 2D+time respiratory phantom, and demonstrated the feasibility using a simple filtered backprojection algorithm along with the proposed motion suppression technique to reconstruct 2D+time images. However, the concepts can be expanded to 3D+time (i.e. 4D) datasets both using conventional CT or CBCT acquisition. The core of the technique rests on separating out the motion components from individual projection data by selecting the minimum intensity between projections and digitally reconstructed projections. This information is then later added back into the final images reconstructed at each individual phase to suppress motion artifacts. Results: Initial results on a 2D+time simulated phantom proves that severe under-sampling artifacts can be eliminated and individual phase images can be reconstructed with high fidelity when compared to standard under-sampled filtered backprojection. Conclusion: The technique is successfully demonstrated using simple backprojection, and hence computational complexity is comparable to standard backprojection (for CT) or FDK (for CBCT). Computational complexity will be significantly better when compared to traditional iterative methods that use regularization and penalty functions. NIH R01CA133539

Purpose: To commission electron Monte Carlo (eMC) algorithm in Eclipse Treatment Planning System (TPS) for TrueBeam Linacs, including the evaluation of dose calculation accuracy for small fields and oblique beams and comparison with the existing eMC model for Clinacs. Methods: Electron beam percent-depth-dose (PDDs) and profiles with and without applicators, as well as output factors, were measured from two Varian TrueBeam machines. Measured data were compared against the Varian TrueBeam Representative Beam Data (VTBRBD). The selected data set was transferred into Eclipse for beam configuration. Dose calculation accuracy from eMC was evaluated for open fields, small cut-out fields, and oblique beams at different incident angles. The TrueBeam data was compared to the existing Clinac data and eMC model to evaluate the differences among Linac types. Results: Our measured data indicated that electron beam PDDs from our TrueBeam machines are well matched to those from our Varian Clinac machines, but in-air profiles, cone factors and open-filed output factors are significantly different. The data from our two TrueBeam machines were well represented by the VTBRBD. Variations of TrueBeam PDDs and profiles were within the 2% /2mm criteria for all energies, and the output factors for fields with and without applicators all agreemore » within 2%. Obliquity factor for two clinically relevant applicator sizes (10×10 and 15×15 cm{sup 2}) and three oblique angles (15, 30, and 45 degree) were measured for nominal R100, R90, and R80 of each electron beam energy. Comparisons of calculations using eMC of obliquity factors and cut-out factors versus measurements will be presented. Conclusion: eMC algorithm in Eclipse TPS can be configured using the VTBRBD. Significant differences between TrueBeam and Clinacs were found in in-air profiles and open field output factors. The accuracy of the eMC algorithm was evaluated for a wide range of cut-out factors and oblique incidence.« less

Purpose: To quantitatively evaluate the inter-fractional variations in ITVs over the course of treatment using multiple 4DCTs and CBCTs for patients treated with SBRT for lung targets. To compare using ITV generated from all phases of a 4D or average intensity projection (ITV-AIP) as reference contours for image guidance. Methods: An IRB approved study included five patients who received SBRT for a lung target using daily image guidance with a CBCT. Three 4DCTs were acquired before (simulation), during and at the end of treatment. ITV-4Ds (on all 10 phases) and ITV-CBCTs were contoured manually. All CT images and contours were registered to the reference image - AIP1 from 4D1. The registration consists of intensity-based image registration followed by manual soft-tissue alignment, where physicians manually make shift to ensure that the tumor seen on CBCT is encompassed in ITV-4D1 or ITV-AIP1. The relationship between two ITVs was quantified by centroid distance, volume inclusion, and volume overlap (Dice Coefficient). Results: The average volume for ITV-4D, ITV-AIP, and ITV-CBCT were 23.1, 17.2, and 14.1 cm3, respectively. When using ITV-4D1 as a reference contour in the manual soft-tissue alignment, an average of 93% of ITV-CBCTs was included in ITV-4D1, with a centroid distance of 2.0 mm. Only 81% of ITV-4Ds were included in ITV-4D1. The Dice Coefficient between ITV-4D1 and ITV-4Ds was only 0.68, suggesting that the two ITVs were not highly overlapping. For all comparisons, the results were similar using either ITV-AIP1 or ITV-4D1 as a reference contour in the manual soft-tissue alignment. Conclusions: There were noteworthy inter-fractional variations in ITV-4Ds. Compared to ITV-4D, ITV-AIP was much similar to ITV-CBCT. However, there was no significant difference between using ITV-4D1 and using ITV-AIP1 as a reference contour for image guidance. Further dosimetric studies are required to evaluate actual tumor coverage. This study was supported in part by Philips Healthcare.

Purpose: We have developed a new technique called dose‐rate‐modulated tracking (DRMT) for closed‐feedback real‐time tumor tracking by changing the dose rate. Method and Materials: DRMT uses pre‐programmed MLC sequences that are generated using measured data for tumor motion obtained at an earlier time. Since the leaves move, their positions are programmed with schedules that are a function of the accumulated dose. The leaf trajectory on the time axis then can be changed during treatment by changing the dose rate. DRMT changes the dose rate to minimize the discrepancy between the scheduled MLC position and the target position or breathing signal on the day of treatment. If the monitored breathing is slower than that observed at simulation, the dose rate is lowered to slow down the movement of the MLC and vice versa, thereby maintaining synchrony. DRMT tumor‐tracking power was tested with sinusoidalbreathing functions and patient‐breathing signals (RPM, Varian). The tracking error (2σ) for each breathing signal was derived as a function of the system reaction time and the dose‐rate correction period. Results and Conclusions: DRMT simulation showed that for the sinusoidal‐breathing signal with 2‐cm peak‐to‐peak tumor motion, less than 0.2 cm tracking error was obtained if the system can react to a detected mismatch in less than 0.3 s and the dose rate can be adjusted in 0.36 s. Fourteen out of 26 patients were eligible for DRMT. The selection criteria were: (1) peak‐to‐peak breathing motion greater than 0.5 cm and (2) amplitude variation less than 20 % during the DRMT simulation. The tracking error for the patient data is expressed as a percentage of the motion amplitude. A tracking error of less than 20 % was achieved for 13 out of the 14 patients if the system can respond in 0 s.

Purpose: To identify clinically‐driven MLC performance criteria for non‐split step‐and‐shoot IMRT fields by quantifying the dosimetric impact of Varian MLC leaf abutment miscalibration on the delivered dose. The maximum tolerable leaf gap that can be used as a criterion for MLC leaf gap QA will be determined. Methods: Ten IMRT plans with large (X > 15 cm), non‐split fields were selected, 5 whole pelvis and 5 head and neck. All were planned using Pinnacle v9.0. The delivered dose for each plan was calculated using Monte Carlo simulations assuming leaf abutment gaps from 0 to 3 mm, in 0.5 mm increments. All calculations were done with 1 mm resolution. Three dimensional dose difference maps (ΔD) were calculated by subtracting the 0 mm dose matrix from each non‐zero one. Cumulative ΔD histograms were constructed to determine the maximum tolerable gap where 5% volume received ΔD >= 3%. Results: The gap where % volume received ΔD >= 3% ranged from 0.7 to 2.2 mm (average 1.0 mm). For 8/10 plans this gap was >= 1.0 mm. The head and neck plans showed more sensitivity to gaps than the whole pelvis plans (p = 0.08). Increasing gap size did not adversely affect target coverage. Maximum OAR dose increased with increasing gap; a 3 mm gap led to a 4% dose increase. Conclusion: Non‐split fields should be used for IMRT with caution. This method of treatment is acceptable as long as MLC leaves abut within 0.5 mm. Standard IMRT QA using a 4 mm calculation grid and gamma of 3%/3mm is not sensitive enough to detect dose differences from leaf abutment miscalibration. If treating this way, leaf abutment should be checked as part of weekly QA with a 0.5 mm criterion in addition to the TG‐142 recommended MLC QA.

Purpose: The GammaPodTM stereotactic radiotherapy device incorporates a novel stereotactic immobilization system and a fiducial based localization frame for the breast. We present the localization accuracy achieved with this stereotactic system through an IRB approved prospective protocol. Methods: GammaPodTM target localization is achieved using a dual layered immobilization breast cup under a negative pressure which has a built-in stereotactic frame. Patient CT scans were attained using a pivoted patient loader which automatically lowers the patient from a vertical position to a horizontal prone-lying position. We then performed two CT scans 30 minutes apart, simulating the pre-treatment imaging and treatment sessions. The patient is disconnected from the patient loader after the first scan, but the immobilization cup remains on the breast under 150 mmHg of negative pressure until the end of the second scan. Evaluation of target localization was carried out via previously implanted surgical clips (two to five) and superficially positioned markers (three to seven). Scans were performed with 1 mm CT slice thickness. Results: The first 9 patients' results are presented with a total of 25 patients expected. Two patients were excluded due to improper fitting of the breast cup and patient comfort; one patient was excluded due to loss of breast cup pressure between the two scans. In the remaining six patients, the spatial reproducibility of internal markers was found to be 1.2 ± 1.0 mm (range 0.3−3.7) in between the two scans while the external marker positions differed by 0.9 ± 0.9 mm (range 0.1−3.9). The intrinsic accuracy of the stereotactic localization system found to be 0.1 ± 0.1 mm. Conclusion: The GammaPodTM target localization system is a unique approach to breast immobilization employing stereotactic principles. Preliminary results of this study suggest that the overall localization uncertainty of this system is 2.8 mm (95% CL). C. Yu and W. Regine have financial interests in Xcision Medical Systems.

Purpose: Due to the nature of the beam fluence of the flattening‐filter free (FFF) mode from Varian TrueBeam linear accelerator, the tissue‐maximum ratio (TMR) conversion formalism from percent‐depth dose (PDD) may not be appropriate at large field sizes and large depths. The purpose of this work was to establish an analytical correction formalism which accounts for the differences in FFF‐mode beam profiles that corrects for small differences in TMR between the calculations and measurements. Methods: Two FFF‐mode beams (6‐MV and 10‐MV) were investigated using Varian TrueBeam linear accelerator. Initially, the analytical correction formalism starts out with the TMR conversion equation which is the product of the PDD, inverse square, and phantom scatter ratio. A ratio of two off‐axis ratios (OARs) at the surface and depth of interest is determined for a given field size. This OAR ratio is multiplied with the phantom scatter term in the TMR conversion equation to account for small differences in TMR values at large field sizes (>= 15 × 15 cm2) and depths (>= 15 cm). Results: For 10‐MV FFF‐mode, the difference between the measured and the standard TMR formalism ranged from ‐ 3.2% to ‐ 0.9% (mean = ‐ 1.8% and 1‐standard deviation = 0.7%) for field sizes >= 15 × 15 cm2 and >= 15 cm depth. When the proposed correction formalism was applied to the TMR conversion formalism, the difference between the measured and the corrected TMRs for field sizes >= 15 × 15 cm2 and >= 15 cm depth ranged from −0.7% to −0.2% (mean = −0.4% and 1‐standard deviation = 0.1%). Similar relationships were observed for 6‐MV FFF‐mode. Conclusion: The analytical correction formalism which accounts for the differences in the FFF‐mode beam profiles has been shown to correct for the small differences in TMR that were observed without the correction.

Purpose: to present a method to generate retrospectively a pre‐selected motion phase cone‐beam CT image from the full motion cone‐beam CT acquired at standard rotation speed. Materials and Methods: the method consists of subtracting from the full CBCT all of the undesired motion phases and obtain a motion de‐blurred single‐phase CBCT image CTnm. This is accomplished by means of a subtraction CBCT (CTs) added to the full CBCT (CTm). CTs is obtained as follows: i) we sort the CBCT projections into each phase bin k according to the known motion trace; ii) compute the digitally reconstructed radiographs (DRR) from the full reconstruction at the gantry angles correlated to bin k; iii) generate a set of subtraction projections for phase k by subtracting the DRR from the original projection at the same gantry angle; v) reconstruct CTs from this single‐phase subtraction projection set; vi) add CTs to CTm to obtain CTnm. A qualitative assessment of the image quality improvement was performed using a simple shapes phantom. To quantitatively study the method, we used the image quality modules of a Catphan phantom, evaluated in a static scan and under anterior‐posterior motion. Results: The single phase CBCT reconstruction generated by our method successfully isolates the desired motion phase from the full motion CBCT, effectively reducing motion blur. It also shows improved image quality, with reduced streak artifacts with respect to the reconstructions from unprocessed phase‐sorted projections only. Conclusions: A novel CBCT motion de‐blurring algorithm has been developed and tested with phantom data. The algorithm allows to improve visualization of a single phase motion extracted from a standard CBCT dataset. NIH R01CA133539

A dedicated stereotactic irradiation device, the GammaPodTM, was developed to treat early stage breast cancer. This study presents the first description of the dosimetric and geometric characteristics from the prototype unit.The GammaPod stereotactic radiotherapy device is an assembly of a hemi-spherical source carrier containing 36 Co-60 sources, a tungsten collimator, a dynamically controlled treatment table and a breast immobilization cup embedded with a stereotactic coordinate system. The source carrier and the variable-size collimator rotate synchronously to form 36 non-coplanar, concentric arcs focused at the isocenter. The treatment table enables motion in three dimensions facilitating continuous dose painting in comparison to a sphere packing approach. Geometric and dosimetric evaluations and a method for absorbed dose calibration are provided. Dosimetric verifications of the dynamically delivered plans are performed for eight patients in hypothetical pre-op, post-op and dose painting treatment scenarios.Loaded with a cumulative activity of 4320 Ci, the GammaPod unit delivers 5.31 Gy/min at the isocenter. Due to non-coplanar beam arrangement and dynamic dose shaping features, the GammaPod delivers uniform doses to the targets with excellent conformity. The spatial accuracy of the device is less than 1 mm. Single shot profiles with the 25 mm collimator are measured with radiochromic film and found to be in good agreement with respect to the MC based calculations (congruence of FWHM less than 1 mm). Dosimetric verifications corresponding to all treatment plans corresponding to three target scenarios for each of the eight patients demonstrated Gamma index pass rates greater than 97%.The first description of the dosimetric and geometric evaluation of the GammaPod was performed. The observed level of agreement between the treatment planning system calculations and dosimetric measurements has confirmed that the system can deliver highly complex treatment plans with remarkable geometric and dosimetric accuracy. C Yu and J Zhang have commercial affiliations with Xcision Medical Systems.

Purpose: In the process of measuring and validating fundamental dosimetry data prior to the clinical use of a treatment unit, it is prudent to compare measurements with previously published equivalent data. During the commissioning of an accelerator with flattening filter free (FFF) photon beams (Varian True Beam 6 MV FFF and 10 MV FFF) we compared measured Phantom Scatter Factors (Sp) with the Normalized Peak Scatter Factors (NPSFs) from the British Journal of Radiology Supplement 25 (BJR #25). The purpose of this work was to determine whether the energy independent NPSFs BJR#25 are valid for comparison with FFF photon beams. Methods: All measurements were performed using a Varian TrueBeam linear accelerator with photon energies of 6‐MV, 6‐MV FFF and 10‐MV FFF modes. For all measurements, a Scanditronix CC04 ionization chamber was used. Both water and in‐air measurements were made to obtain NPSFs normalized to the 10 × 10 cm2field size. For measurements in water, the chamber was positioned at 100 cm source‐to‐axis distance at the depth of dose maximum. For in‐air measurements, the chamber was positioned at 100 cm source‐to‐axis distance with appropriate build‐up cap. From BJR #25, NPSFs were obtained for comparison with the measurements. Results: The NPSF agreement between the 6‐MV and 6‐MV FFF with the BJR#25 were all within ±0.5%. The agreement ranged from 0.996 to 1.004 and 0.995 to 1.002 for 6‐MV and 6‐MV FFF, respectively. We also found that 10‐MV FFF showed very similar trend. Conclusions: The scatter factors reported in BJR #25 are valid for comparison for 6‐MV, 6‐MV FFF, and 10‐MV. Additional investigation is needed to further understand the dosimetric characteristics of FFF mode.