H Saleh

Regulatory evaluation of prostate volume implants: pitfalls of a retrospective assessment.

PURPOSE Evaluate for regulatory compliance the prostate implants from the Philadelphia Veterans Medical Center applying both an activity-based and volume-corrected D(90) (the maximum dose delivered to 90% of the prostate volume) metrics. METHODS AND MATERIALS Dosimetry from 107 prostate implants performed at the Philadelphia Veterans Medical Center used immediate postprocedural CT image sets. D(90) values were adjusted for volume differences from planning volumes. Medical events (MEs) determined from the volume-corrected data were compared with an activity-based metric. RESULTS Examination of images using original and third-party reviewed prostate contours revealed 56 and 62 cases with D(90) values <80% of the prescription dose, respectively. Because postprocedural prostate volumes were on average 55.7% larger than the planned volume, clinical nomogram-based doses using the implanted activity and actual volumes found 34-47 implants failing to achieve doses greater than 80% of the prescription dose. Volume correction identified 20 MEs, 9 cases with D(90) values within 4% of the ME threshold and 11 significantly inferior cases with median D(90) values <52% of the prescribed dose. Eleven implants also had 20% or more seeds beyond the treatment site according to an activity metric recommended by the VHA Blue Ribbon Panel. Ten of these 11 cases were also identified by volume-corrected D(90) metric. The remaining 96 cases, however, had 95% (±6%) of seeds placed within the treatment site. CONCLUSIONS Of the cases reported to the United States Nuclear Regulatory Commission (NRC) on the basis of Day-1 D(90) values, many appear to have been acceptable implants relative to standard-of-practice clinical criteria. The activity-based dose metric, endorsed by the NRC Advisory Committee on the Medical Uses of Isotopes in 2005 and recommended by the VHA Blue Ribbon Panel for Prostate Brachytherapy yields a more robust determination of ME for this population of implants.

Comparisons of multiple automated anatomy-based image-guidance methods for patient setup before head/neck external beam radiotherapy.

The purpose was to assess the variability in automated translational head/neck setup corrections computed from several different imaging modalities and rigid registration methods using patient anatomy. Shifts were calculated using three commercial and one in‐house automated rigid registration methods for nine head/neck patients who were imaged with three different image‐guidance systems. The mean difference between the daily isocenter shifts determined by the four methods ranged from 2.8 to 12.5 mm for all of the test cases. These differences are much greater than the variability observed for a rigid imaging phantom. Image‐guided setup procedures have an uncertainty that depends on the imaging modality, the registration algorithm, the image resolution and the image content. In the absence of an absolute ground truth, the variation in the shifts calculated by several different methods provides a useful estimate of that uncertainty. PACS number: 87,55,km, 87.57.nj, 87.59.‐e, 87.59.bd

Time-of-flight calibration of a 6Li glass epithermal neutron detector

The curing of Portland cement concrete involves the conversion of water from a free to a bound state. The process can be monitored nondestructively by measuring the shift in the neutron energy spectrum in the epithermal range (0.025-1 eV). A tuned array of 6Li glass detectors has been constructed with varying efficiencies over the epithermal energy range. To determine the efficiency of each detector as a function of neutron energy, it is necessary to calibrate it against a reference neutron spectrum. This was accomplished using a time-of-flight approach with a pulsed neutron beam produced at the Gaerttner LINAC Laboratory at Rensselaer Polytechnic Institute. With a neutron flight path of 25 m it was possible to determine the neutron detector efficiencies to an energy resolution of 11 microeV. The data showed good agreement with the detector design calculations.

SU‐GG‐J‐31: Simultaneous Estimation of Beam Geometry and Radiation/Imaging Isocenter Coincidence in Cone‐Beam CT‐Guided Radiation Therapy

Purpose: To simultaneously characterize the geometric pose of an on‐board CBCTimaging system and measure the relative positions of the imaging and treatment machine radiation isocenters. Method and Materials: The method is based on an existing technique (Cho, et al, Med Phys 32(4), 968), which utilizes CBCT projections of a fiducial phantom to calculate the source and detector positions and orientations as a function of gantry angle. The algorithm was modified to localize the phantom pose relative to the kV source trajectory, and to transform beam geometry parameters from the phantom to imaging isocenter coordinate system. The method was applied to both CBCT and MV EPID projections acquired during the same imaging session without moving the phantom. The calculated positions of the phantom relative to the CBCT and EPID projection isocenters provide the relative displacement of the imaging and treatment isocenters. Two additional CBCTimaging experiments were then performed with the phantom displaced known distances. Results: Calculated CBCT geometric parameters were consistent over the three experiments, even with the phantom displaced from the machine isocenter by 10mm or misaligned with the gantry rotation axis by 1°. The CBCT beam parameters were close to nominal with small variations with gantry angle, with the exception of a 1 mm detector offset parallel to the gantry axis. Displacement of the CBCTimaging isocenter from the treatment isocenter was found to be (0.07, 1.24, 0.05 mm) in the (L‐R, A‐P, S‐I) direction. Conclusion: The methods output can assist in mitigation of geometric distortion‐related CBCTimage artifacts and simultaneously provides the imaging‐to‐treatment isocenter offset. Supported by NCI Grant P01 CA 116602

SU‐FF‐J‐120: Assessing Intrafraction Motion for Cranial Tumor Patients Treated with Hypofractionated Radiation Therapy Using Repeated Imaging

Purpose: To assess intra‐fraction motion of patients treated for braincancer using hypofractionated treatment, to optimize the frequency of patient imaging, and to determine whether treatment margins are appropriate or not. Material and methods: Intra‐fraction patient displacement data for twenty braincancer patients were analyzed retrospectively. Patient imaging for setup correction is done at three stages. The first set of images was taken before the patient treatment. The second set of images was taken after one third of the treatment is delivered, and the third imaging set was taken after two thirds of the treatment is delivered. During each imaging stage, two stereoscopic x‐ray images are acquired using the ExacTrac imaging system. All images were automatically fused with digitally reconstructed radiographs (DRRs) determine patient displacement from the planning isocenter. Displacements were quantified in terms of position changes as a function of time between each two consecutive sets of stereotactic images.Results: The average patient displacements were 0.86 ± .66 mm, 1.15 ± 1.08 mm, and 0.86 ± 0.79 mm in the lateral, longitudinal and vertical directions respectively. During the first one third of the treatment, the average patient displacements were: 0.68 ± 0.62 mm, 0.83 ± 0.66 mm, and 0.65 ±0.78 mm. During the second third of the treatment, the average patient displacements were: 0.57 ± 0.64 m, 0.92 ± 1.08 mm, and 0.69 ±0.79 mm in the lateral, longitudinal, and vertical directions respectively. Conclusions: The majority of positioning errors in this study are within the safety margin of 3 mm. Without patient displacement correction, higher margins may be necessary. Because the average patient displacement during the first one third of the treatment is comparable to the patient displacement during the second third of the treatment, it may not be necessary to increase the frequency of patient imaging.

SU‐GG‐T‐148: Quantification of IMRT Patient Dose Deviations Due to Daily MLC‐Leaf Positional Variations

Purpose: To develop a tool utilizing MLC‐log files and Monte Carlo (MC)dose recalculation that will allow cumulative deviations of IMRT patient plans to be assessed over the treatment course by taking into account the possible inaccuracy in the IMRT delivery. Method and Materials: Prostate, head and neck and lungIMRT patients were selected. The delivered MLC positions during the IMRT delivery were recorded in the form of MLC‐log files which were converted to MLC leaf‐sequence files. These files were then transferred back to the TPS and were attached to the original planned IMRT fields to regenerate the delivered dose distribution. Delivered fractional dose to each patient was then re‐produced using MC calculation. The recalculated fractional doses calculated via MC, which utilized the delivered MLC leaf‐sequences, were then combined to produce correct cumulative dose distribution received by the patient during treatment. Both delivered fractional and cumulative dose distributions and corresponding Dose‐Volume statistics were then compared with the ones obtained from the original IMRT plan. Results: For H&N IMRT case, the fractional changes in delivered dose‐volume indices were <1%. The changes in critical structure doses were negligible. For lungIMRT case, the differences in all dose‐volume parameters with MC re‐calculation using the delivered MLC‐sequences and the original planned MCdose calculation were less than 1.1% for PTV and all critical structures, except for the esophagus where differences in D30 were 2.4%. The similar results have been obtained for the prostate case in the study. Conclusion: The utilization of Monte Carlo for the recalculation of actual fractional doses delivered to the patients using the delivered MLC leaf sequence files for each IMRT field provides a powerful tool for offline QA of IMRT patients and can also be utilized for the verification of adaptive radiotherapy. This work supported by NIH grant P01CA11602.

SU‐GG‐J‐37: Assessing Image Guided Radiation Therapy Targeting Accuracy: Coincidence Of Imaging System Isocenter With Treatment Machine Mechanical And Radiation Isocenters

Purpose: The purpose of this study is to introduce a method for quantifying the accuracy of linac and imaging system isocenters and demonstrate its use by measuring the targeting of two kilovoltage (kV) imaging systems, and one gantry mounted megavoltage Electronic Portal Imaging Device(EPID). One is the kV imaging system (OBI) integrated with the linear accelerator which supports radiographic and cone beam CTimaging. The other system is the BrainLab Exac Trac stand alone fixed to the treatment room ceiling and floor. Method and Materials: A cylindrical Lucite phantom was designed and fabricated to perform this study. The cylindrical phantom is 20 cm in diameter and 20 cm long. The phantom contains 13 radio‐opaque fiducial markers. A 5 mm spherical marker is positioned at the center of the cylinder. The phantom was imaged and the CTdata sets were imported to two planning systems. Treatment beams were placed and both 3D data sets and isocenter information were transferred to the Linac delivery and imaging workstations. For each imaging system, phantom was initially positioned according to machine cross‐hair. Then images were acquired and measured shifts were applied to precisely position the phantom. Then, Winston‐Lutz test was performed for different gantry and collimator configuration. The radiation beam was collimated by a tray‐mounted 30 mm diameter circular collimator.Results: The radius of spherical volume in which all isocenter intersect or lie was measured. For EPID, OBI, CBCT, and Exac Trac, the radius is: 0.54 mm, 1.4 mm, 1.7 mm and 1.46 mm respectively. From the Winston‐Lutz analysis, the average isocenter deviation from all angles is 1.18 ± 0.28 mm. Conclusion: The error in mechanical, radiation and imaging isocenter is about one millimeter. Regardless what imaging system is used for patient setup, it is important to incorporate this systematic error in any planning margins.