Encrypted login | home

Program Information

Development of TG-142 Imaging QA Baselines and Imaging Performance Trending and Analysis Using a QA Data Repository for Nine Beam Matched and Dosimetrically Equivalent TrueBeams Over Two Years of Operation

N Gupta

A Ayan , J Woollard , D DiCostanzo , S Grzetic , J Hessler , N Gupta*, Ohio State University Medical Center, Columbus, OH


WE-F-205-1 (Wednesday, August 2, 2017) 1:45 PM - 3:45 PM Room: 205

Purpose: The purpose of this work was to develop TG142 QA baselines and tolerances for imaging QA and to mine and analyze two years of imaging QA data for nine TrueBeams.

Methods: Beginning December 2014 we have clinically commissioned nine TrueBeams within our institution and implemented common TG-142 baselines for monthly and annual QA as well as identical QA processes. TG-142 does not provide baselines for many of the planar and CBCT imaging parameters, but recommends developing baselines based on acceptance data. Standardized imaging protocols, testing and image processing were employed initially to analyze and develop imaging baselines and common pass/fail criteria across all machines. All QA data were then automatically stored in an SQL database and dashboards were implemented to provide a dynamic data trending interface and statistical analysis.

Results: Analysis of our initial standardized imaging QA data allowed us to create baseline and pass/fail criteria for each imaging protocol for our specific test tools, which we have successfully used over the past two years. For example, our measured HU value for LDPE over all 9 linacs over 2 years show a mean +SD of -96.28 +6.11 for the Head protocol, and -93.99 +6.82 for the Pelvis protocol, with 3SD ranges well below the +50HU specification by Varian. The developed dashboards have provided tools to dissect the data, observe the trends in different parameters, and allowed flagging of outliers for early corrective actions, and introduce statistical process control into our QA process.

Conclusion: Our unique opportunity of standardizing QA processes and methods and mining QA data has allowed us to gain valuable insights and establish, and consistently exceed, tolerances for imaging parameters that TG-142 does not provide explicit values for and are otherwise very difficult to otherwise establish at most institutions.

Contact Email: