Streamlining and Automating Dosimetric Analysis for Treatment Planning System and Linac QA
N Ozturk1,2*, B Smith2, K Ahn2, B Aydogan1,2, (1) University of Chicago, Chicago, IL, (2) University of Illinois at Chicago, Chicago, ILSU-E-T-10 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: To streamline and automate analysis of dosimetry checks for treatment planning system (TPS) commissioning, and TPS and linac QA, through scripts.
Methods: A dosimetric analysis tool was developed using scripts written in Python language. A "standard" dosimetry dataset, ideally data measured for commissioning and/or verification of linac and TPS using a water scanning system, is generated and structured into an Excel spreadsheet. Annual dosimetry measurements, made with water a scanning system, are converted to the same format; measurements with a 2D ion chamber array (Matrixx) are kept in their original format in a separate folder. Plans for the same setup geometries are generated, or existing plans recalculated, and associated 3D dose files are exported in DICOM format into proper folders. Python scripts cycle through the planned DICOM dose files, match the information (beam energy, field size, SSD, depth) in the DICOM header with the data in the Excel files, or in the Matrixx data files, extract relevant dosimetric data from the 3D dose files (PDDs, profiles, output factors), compare to measurement using an in-house developed gamma analysis routine and display the results in a pdf file (profiles, PDDs, planar dose distributions) or tabulate in an Excel sheet (output factors, absolute doses for reference geometries, wedge factors). For points failing the gamma analysis default passing criteria, an optimization is done to iteratively find at which tolerance levels these points pass for further evaluation.
Results: Using our dosimetric analysis software, a test analysis of planned and measured dosimetric data for TPS commissioning of photon and electron beams was completed in about half hour, once data were converted into the proper format.
Conclusion: Scripts written in Python facilitate more comprehensive analysis of dosimetric data for TPS and linac commissioning/QA. Streamlining the data analysis through scripts increases efficiency by reducing repetitive manual tasks.