LAUSR

PURPOSE To establish a method for evaluating the effect of scanning ion beam delivery error on three-dimensional (3D) dose reconstructed on patients' CT based on log file. MATERIALS AND METHODS This study used the MATLAB program to reconstruct the 3D dose on the patient's CT based on the log file (Doselog ) for treatment delivery accuracy check. In addition, differences between the parameters in the log file and the treatment plan, such as the spot position, spot size and particle number, were analyzed, as well as their effects on the dose distribution. The accuracy of the dose reconstruction algorithm was verified by comparing dose from TPS (DoseTPS ) and the dose recalculated based on the treatment plan (Doserec ). Twenty treatment plans of ten prostate cancer patients received carbon ion therapy, and their corresponding 160 log files were selected for analysis and treatment delivery accuracy check. The regions with dose higher than 10% of the maximum dose were selected and 2mm/2% criteria was used for global gamma analysis. Multiple linear regression was use to evaluate the relation between dose deviation and delivery errors. RESULTS For the algorithm accuracy verification, the mean relative dose difference is 1.02% ± 0.12%. For prostate cancer patients treated in our facility using carbon ion radiotherapy, the average passing rate of the gamma analysis between the Doselog and the DoseTPS was 95.3%. The dose deviation caused by the difference in the spot position and the number of particles was smaller than that caused by spot size deviation. CONCLUSION This study established a 3D dose verification method based on log files to evaluate the accuracy of daily delivered treatment dose. In our facility, the daily delivered dose accuracy of carbon ion therapy for prostate cancer was mainly affected by the spot size deviation in terms of the machine delivery part. This article is protected by copyright. All rights reserved.