Fifty-seventh annual meeting of the American association of physicists in medicine
TH-EF-BRD-08: A Retrospective Statistical Analysis of Consistencies in Treatment Plan Parameters by the Treatment Site and Modality
Simple rules and common senses based on prior knowledge could be powerful tools for medical physicists to identify and prevent severe errors in radiation therapy (RT) treatment plans. However, rules, e.g. the acceptable MU/cGy ratio for different treatment sites, are not yet available quantitatively and therefore cannot be utilized effectively. In this study, we have retrospectively analyzed all the EBRT treatment plans of the past 8 years at authors’ institution. The goal is to better understand the most common treatment plan data statistics, involving MU/cGy ratio, numbers of beams and segments, and ranges of gantry angles for lateral targets, in order to quantitatively define the common sense-based rules that could be used to guide manual or automatic physics checks.
15844 previous treatment plans were obtained from Mosaiq. Plan data were extensively processed including 1) reducing data size, 2) removing empty and unapproved site and fields, 3) deriving common site names from names of prescriptions and beams that were spelled and misspelled extremely diversely, and 4) removing the plans not covered by this study, e.g. HDR plans. Processed data was then statistically analyzed using multivariate regression algorithms to quantitatively investigate the relationships between the input parameters, i.e. treatment site, technique, modality, and laterality, and the output parameters, i.e. MU/cGy radio, the total number of treatment beams and segments, and the beam gantry angles (only for lateral targets).
Mean values and standard deviations of the output parameters were calculated and tabulated per treatment site, technique and modality. Beam gantry angle range and mean gantry angle were analyzed per site for plans treating lateral targets.
The results of this study, as the statistics of the major treatment plan parameters, could be useful to quantitatively evaluate the consistency of new treatment plans in either manual or automatic plan checking procedures.