Evaluation of MLC errors of LINAC based on log file

Abstract

This study aimed to investigate the Multi Leaf Collimator (MLC) quality assurance (QA) and fluence monitor unit (MU) reconstruction based on log files of Varian Linear Accelerator. The graphical user interface (GUI) was developed using MATLAB to analyze the Log files. The MLC QA was analyzed using the error root mean square (RMS) of leaf parameters. Meanwhile, the fluence MU were reconstructed based on the information of leaf position, jaw position, and fractional MU for each 50 ms. The gamma index evaluation between actual and expected fluence MU was calculated to analyze the segment potential error caused by the MLC position. The average of mean error RMS leaves for Bank A and Bank B are 0.213 ± 0.098 mm and 0.207 ± 0.073 mm, whereas the average of maximum error RMS leaves for Bank A and Bank B are 0.491 ± 0.056 mm and 0.48 ± 0.061 mm, respectively. The mean gamma index results of 2%, 2mm and 3%, 3mm criteria between actual and expected fluence MU for the composite beam for all case are 98.2 ± 1.0% and 99.4 ± 0.3%, respectively. The error of MLC based on the log file analysis is within the tolerance level. The results showed that parameters irradiation such as MLC position, jaw position and fractional MU proper with planning.