A Clinical Evaluation of An Open-Source Toolkit for TG-142 Based Monthly Imaging Quality Assurance Via Comparison with Commercial Software
S Jayarathna1*, S Howard2, D Johnson1, (1) The University of Kansas Medical Center, Kansas City, KS, (2) The University of Kansas Cancer Center, Overland Park, KS
Presentations
PO-GePV-M-59 (Sunday, 7/25/2021) [Eastern Time (GMT-4)]
Purpose: To assess and evaluate the quantitative differences of imaging parameters for TG-142 monthly quality assurance (QA) of MV- EPID (electron portal imaging device) and kV-CBCT (cone beam computed tomography) image post-processing utilizing an open-source toolkit Pylinac (PL) in comparison with the commercially available software PIPSpro (PP).
Methods: The QC3 MV imaging phantom (Standard Imaging) and Catphan 504 CT phantom (The Phantom Lab) were imaged upon multiple Varian liner accelerators as specified by the manufacturer. Modular transfer functions (MTF) were measured using PL and PP respectively. The Catphan phantom was imaged via half-fan CBCT and analyzed in PL and PP to evaluate Hounsfield Unit (HU) uniformity, HU linearity, and spatial resolution. The results of each software system were then compared against each other and against the baseline previously established at our institution.
Results: For MV imaging the MTF value differences obtained by PL were found to be 14 % higher than those established in PP for all the frequencies considered. PL based kV-CBCT testing revealed HU uniformity values greater by ~ 9% for all 5 regions of interests. For HU linearity, a difference of < 2% was found, both PL/PI reporting HU greater than standard values. kV-CBCT MTF comparisons demonstrate that PL values differ by < 2% for frequencies < 0.3 lp/mm, deviating more than 5% beyond that.
Conclusion: PL and PP measured imaging parameters demonstrated similar results. PL can allow for quick and reliable estimation of the TG-142 related image post-processing parameters. The presented results warrant further investigation into other systems and phantom combinations to fully incorporate the PL in a clinical environment. Nevertheless, the use of the software to establish and monitor baseline conditions as outlined by TG142 and manufacturers maybe a cost-effective solution for further automation in the field.
ePosters
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email