No abstract

Novalis Tx ExacTrac X‐ray system has the 6D adjustment ability for patient setup. Limited studies exist about the setup uncertainty with ExacTrac X‐ray system for IMRT prostate treatment with fiducial markers implanted. The purpose of this study is to investigate the marker‐based prostate IMRT treatment setup uncertainty using ExacTrac 6D IGRT ability for patient setup. Forty‐three patients with prostate cancers and markers implanted have been treated on the Novalis Tx machine. The ExacTrac X‐ray system has been used for the patient pretreatment setup and intratreatment verification. In total, the shifts data for 1261 fractions and 3504 correction times (the numbers of X‐ray images were taken from tube 1 and tube 2) have been analyzed. The setup uncertainty has been separated into uncertainties in 6D. Marker matching uncertainty was also analyzed. Correction frequency probability density function was plotted, and the radiation dose for imaging was calculated. The minimum, average, and maximum translation shifts were: −5.12±3.89 mm, 0.20±2.21 mm, and 6.07±4.44 mm, respectively, in the lateral direction; −6.80±3.21 mm, −1.09±2.21 mm, and 3.12±2.62 mm, respectively, in the longitudinal direction; and −7.33±3.46 mm, −0.93±2.70 mm, and 5.93±4.85 mm, respectively, in the vertical direction. The minimum, average, and maximum rotation shifts were: −1.23° ± 1.95°, −0.25° ± 1.30°, and −2.38° ± 2.91°, respectively, along lateral direction; −0.67° ± 0.91°, −0.10° ± 0.61°, and −1.51° ± 2.04°, respectively, along longitudinal direction; and −0.75° ± 1.01°, −0.02° ± 0.50°, and −0.82° ± 1.13°, respectively, along vertical direction. On average, each patient had three correction times during one fraction treatment. The radiation dose is about 3 mSv per fraction. With the ExacTrac 6D X‐ray system, the prostate IMRT treatment with marker implanted can achieve less than 2 mm setup uncertainty in translations, and less than 0.25° in rotations as overall interfraction mean error. The imaging dose is less than kV (CBCT) for setup verification.PACS number: 87.55.km, 87.55.Qr, 87.56.bd

Purpose: Electron Monte Carlo (eMC) and Pencil‐Beam (PB) algorithms are both used clinically for electron beam radiation dosimetry. The purpose of this study is to evaluate the two calculation algorithms with reference to the standard electron beam dosimetry parameters and isodose distributions in practical clinical applications. Methods: eMC and PB algorithms were both commissioned in Eclipse treatment planning system. Standard radiation dosimetry parameters ‐ ‐including percent depth dose, beam profile width, and output factors‐‐were calculated in homogeneous water phantom for various combinations of electron energy and cone applicators. The calculation results were compared with the measured data. The two algorithms were also compared in clinical situation. A phantom simulating chest wall and lung was created and used for dose calculation. A 9 MeV electron beam with a customized cutout on the 20−20 cone was used at fixed SSD of 100cm. The calculation setting for eMC was 1% precision, 1.5‐mm grid, and strong Gaussian smoothing. For PB, 2.5mm grid was used for dose calculation. Results: The difference between measured and calculated depths of R100, R90, R80, and R50 for both algorithms is less than 1mm for all combinations of energies and cone applicators. For the field width and output factors, eMC affords slightly better agreement with measurement than PB does. For practical phantom calculations with both surface irregularity and heterogeneity, eMC shows a more reasonable isodose distribution. Comparing to eMC, PB algorithm tends to overestimate the dose neighboring the air cavity and underestimate the dose past the air cavity in a water phantom. For surface irregularity, eMC and PB show similar results though. Conclusion: eMC and PB algorithms are accurate in the calculation of standard electron dosimetry parameters. Both eMC and PB can handle the surface irregularity very well. eMC is superior to PB in terms of heterogeneity.

Purpose: To evaluate if the addition of dental putty based bite block improves the setup accuracy over the immobilization mask alone in the treatment of brain metastasis using linear accelerator based stereotactic radiosurery. Methods: Localization shift data from on‐board kV CBCT images relative to reference CT images of 16 patients receiving brain stereotactic radiosurgery were retrieved and analyzed for this study. Out of the 16 patients, 8 were treated with immobilization mask alone and the remaining were treated with a combination of dental putty based bite block and immobilization mask. Due to the unavailability of 6D couch in our clinic, we used only the translational registration between CBCT images and reference CT images for couch shifts in patient treatment. The clinical localization shifts were used for data analysis. To compare the two immobilization methods comprehensively, the CBCT images were retrospectively co‐registered with the reference images with six degree of freedom (3 translations and 3 rotations). A paired t‐test was used to evaluate the statistical significance of the two immobilization methods. Results: For the clinical patient treatment, localization shifts from immobilization mask only were 1.5±1.8 mm, 0.1±2.6 mm, and −1.0±2.2 mm, in the vertical, longitudinal, and lateral directions respectively. The addition of dental putty based bite block improves the localization shifts to 0.2±0.2 mm, 0.7±0.6 mm, and 0.9±1.2 mm in the vertical, longitudinal, and lateral directions respectively. Furthermore, the addition of dental putty based bite block reduces significantly the rotational variations of patient head in the immobilization mask. Conclusion: The combination of dental putty based bite block and immobilization mask improves the setup accuracy of linear accelerator based intracranial stereotactic radiosurgery. The formation of a reliable dental putty impression to the upper jaw of each patient is essential for a successful implementation of the procedure.