Rotational artifacts in on-board cone beam computed tomography.

Rotational artifacts in image guidance systems lead to registration errors that affect non-isocentric treatments and dose to off-axis organs-at-risk. This study investigates a rotational artifact in the images acquired with the on-board cone beam computed tomography system XVI (Elekta, Stockholm, Sweden). The goals of the study are to identify the cause of the artifact, to characterize its dependence on other quantities, and to investigate possible solutions. A 30 cm diameter cylindrical phantom is used to acquire clockwise and counterclockwise scans at five speeds (120 to 360 deg min(-1)) on six Elekta linear accelerators from three generations (MLCi, MLCi2 and Agility). Additional scans are acquired with different pulse widths and focal spot sizes for the same mAs. Image quality is evaluated using a common phantom with an in-house three dimensional contrast transfer function attachment. A robust, operator-independent analysis is developed which quantifies rotational artifacts with 0.02° accuracy and imaging system delays with 3 ms accuracy. Results show that the artifact is caused by mislabelling of the projections with a lagging angle due to various imaging system delays. For the most clinically used scan speed (360 deg min(-1)), the artifact is ∼0.5°, which corresponds to ∼0.25° error per scan direction with the standard Elekta procedure for angle calibration. This leads to a 0.5 mm registration error at 11 cm off-center. The artifact increases linearly with scan speed, indicating that the system delay is independent of scan speed. For the most commonly used pulse width of 40 ms, this delay is 34 ± 1 ms, part of which is half the pulse width. Results are consistent among the three linac generations. A software solution that corrects the angles of individual projections is shown to eliminate the rotational error for all scan speeds and directions. Until such a solution is available from the manufacturer, three clinical solutions are presented, which reduce the rotational error without compromising image quality.

Sci-Fri PM: Delivery - 06: A generalized solution to the wide field array calibration method.

Multi detector arrays are commonly used in radiation oncology for IMRT and Linac QA. The calibration of detector arrays is usually based on the wide field calibration technique. Unfortunately small beam changes between measurements will result in large error propagation. The present work introduces a generalized modified version of the wide field calibration method, robust against measurement to measurement variation. Our generalized framework uses an unlimited number of measurement pairs, n geometric positions providing n(n-1)/2 pairs. We solve this large over determined linear system using least squares with gradient method. Measurements were made on an Elekta synergy 6 MV beam with two IBA Matrixx detectors, each containing a 32 × 32 array (1024) of vented pixel ionization chambers. Data acquisition was by the IBA Omnipro Advance software, version 1.2 running in the "ONLINE" cine mode with a 10 sec integration time. Continuous beam sampling (10 seconds long) measured over 10 minutes demonstrated why consistent calibration using the conventional wide field calibration is a challenge. Overall signal changes of 1.6%, flatness changes of 0.3% and the beam symmetry changes of 0.2% over the full 10 minute beam-on time were observed. For the purpose of testing and demonstration of our method, we have chosen to make measurements in 5 geometric orientations relative to the beam, including 1 reference position, 2 rotations and 2 translations. With our method we were able to calibrate all 1024 detectors to better than 0.6% total uncertainty as demonstrated against inter and intra MatriXX comparison.

Dose perturbations by two carbon fiber treatment couches and the ability of a commercial treatment planning system to predict these effects.

PURPOSE: To measure the effect of the treatment couch on dose distributions and to investigate the ability of a modern planning system to accurately model these effects. METHODS: This work measured the dose perturbation at depth and in the dose buildup region when one of two treatment couches, CIVCO (formerly MED-TEC) or Medical Intelligence, was placed between a photon beam source (6, 10, and 18 MV) and the phantom. Beam attenuation was measured in the center of a cylindrical acrylic phantom with a Farmer type ion chamber at multiple gantry angles. Dose buildup was measured in Solid Water with plane parallel ion chambers (NACP-02 and PTW Markus) with the beam normal to both the phantom and couch surfaces. The effective point of measurement method as described [M. R. McEwen et al. "The effective point of measurement of ionization chambers and the build-up anomaly in MV x-ray beams," Med. Phys. 35(3), 950-958 (2008)] was employed to calculate dose in the buildup region. Both experiments were modeled in XiO. Images of the treatment couches were merged with images of the phantoms such that they were included as part of the "patient" image. Dose distributions calculated with superposition and fast superposition algorithms were compared to measurement. RESULTS: The two treatment couches have different radiological signatures and dissimilar water equivalent thicknesses (4.2 vs 6.3 mm.) Maximum attenuation was 7%. Both couches caused significant loss of skin sparing, the worst case showing an increase in surface dose from 17% (no couch) to 88% (with couch). The TPS accurately predicted the surface dose (+/-3%) and the attenuation at depth when the phantom was in contact with the couch. For the open beam the TPS was less successful in the buildup region. CONCLUSIONS: The treatment couch is not radio-transparent. Its presence between the patient and beam source significantly alters dose in the patient. For the most part, a modern treatment planning system can adequately predict the altered dose distribution.

The emerging role of IG-IMRT for palliative radiotherapy: a single-institution experience.

Many modern radiotherapy centers now have image-guided intensity-modulated radiotherapy (IG-IMRT) tools available for clinical use, and the technique offers many options for patients requiring palliative radiotherapy. We describe a single-institution experience with IG-IMRT for short-course palliative radiotherapy, highlighting the unique situations in which the technique can be most effectively used.

Towards an accurate and robust method based on fuzzy logic principles for the reconstruction and quantification of large volumes from MR and CT images.

The authors have previously evaluated a new method of volume reconstruction and quantification from MR images, based on fuzzy logic (FL) principles. The technique is evaluated here for larger and more complex structures by investigating its accuracy and robustness using MR and CT images. Four large (50-71 cm(3)) and complex (e.g. mimicking a prostate) structures were created and imaged on MR and CT scanners, both with increasing slice thickness. Contours were delineated to generate 112 volumes. MR and CT images were processed using the FL method and a "classical" method of reconstruction on research software. In addition, the CT images were also processed on commercial virtual simulation software. Calculated volumes were compared with actual volumes. The mean +/- standard deviation of the relative variations in calculated target volume using the FL method was found to be 4.4%+/-2.8%, whereas with the "classical" method it was 23.7%+/-6% from axial MR images and 23.3%+/-9.8% from CT images. With the "classical" method, the relative variations in calculated volumes rise with increasing slice thickness, and the displayed volumes show deformations in the longitudinal direction. With the FL method, the volume calculation is not sensitive to the slice thickness and so the deformations are minimal. When used with MR images, our FL method of volume reconstruction is accurate and robust with respect to changes in slice thickness. For CT images, the results are encouraging but some work is still needed to improve the accuracy of the FL method.

Poster - Thurs Eve-32: Dose errors related to the treatment couch.

Modern radiotherapy linacs often use carbon fibre for their couch tops due to its radio translucent properties. Beam attenuation by the couches is often ignored during planning and MU calculation. This work examines beam attenuation and loss of "skin sparing" (dose build up region) when various photon beams transit either the MedTec (Siemens) or Medical Intelligence (Elekta) couches. Additionally, measured doses were compared to CMS treatment planning system (XiO version 4.33.02) predictions. We found the two couches to have different structures, resulting in different attenuation signatures as a function of gantry angle. For normal beam incidence the Siemens and Elekta couches had radiological thicknesses of 4.5 mm and 6.0 mm, respectively. For a normal incidence 10×10 cm2 6MV beam the surface dose after couch transmission was 93% vs. 83% for Elekta and Siemens, respectively. Conversely, the increased mass on the lateral edge of the Siemens couch resulted in a maximum attenuation (6 MV 5×5 cm2 beams) of 8% compared to 5% by the Elekta couch. Incorporating the treatment couch as part of the patient planning CT allowed the CMS TPS model to calculate couch attenuation within 1% of measurement, except at the very edge of the Siemens couch, where the attenuation is strongly gantry angle dependent. The CMS beam model was also able to predict the loss of skin sparing within 1%. In conclusion, the two patient couches are different, but both can significantly affect patient dose which can be accounted for in the CMS TPS.