An in-house software program for quantitative image quality evaluation of linac cone-beam CT system

The study presents application of an in-house software program to calculate image quality parameters of kV cone-beam CT (CBCT) system in radiotherapy. The parameters include uniformity, noise, contrast-to-noise ratio, CT number linearity, spatial resolution and geometric distortion as recommended in AAPM TG142 and TG179 quality assurance (QA) programme for image quality performance of kV CBCT. The computer algorithm packaged in the software program is developed using MATLAB (MathWorks, Natick, MA). The algorithm was optimised for two CT image quality phantoms, Catphan 600 (The Phantom Laboratory, Salem, NY) and CIRS 062QA (CIRS, Norfolk, VA). The phantoms were scanned with XVI CBCT system (Elekta AB, Stockholm, Sweden) using an optimised CT imaging protocol. The algorithm measures the image quality metrices from the corresponding image quality test patterns in the reconstructed CT images. The algorithm provides image quality parameters from both phantoms quantitatively. The software program allows the results to be exported and recorded in MS Excel for monitoring the performance trend of the image quality parameters. The software reduces measurement uncertainties from qualitative human observation of the test patterns. (paper)