A Retrospective Study of Commercial Software System EDose in QA Application based on the Electronic Portal Imaging Device

Purpose: To evaluate the functions about the pre-treatment dose verification and, the in vivo dose verification for the commercial software EDose system based on Electronic Portal Imaging Device (EPID) retrospectively and establish the action limit level. Methods: The results of pre-treatment dose verification were compared with 2D array Seven29 and 3Dmap for 50 randomly selected IMRT plans of different lesions. A retrospective analysis was conducted for 287 radiotherapy plans using the EDose in pre-treatment dose verification, including 53 IMRT and 247 RapidArc plans, to establish the action limit level with statistical significance evaluation. 28 head and neck patients with different lesions were selected randomly for studying 3D online dose verification preliminary. Results: For pre-treatment dose verification, 50 plans’ average γ passing rates of the 3%/3mm criterion were > 98% for EDose, Seven29, 3Dmap, and 3%/2mm, 2%/2mm criteria were > 95%, 90%. The average γ mean of the three verification methods were similar for the 3%/3mm criterion (0.35, 0.38, 0.35). Based on the 287 patients’ clinical data, the average γ passing rate was 97.5%, and the recommend clinical action level was established at 92% with a 95% confidence limit. The in vivo results showed that the γ pass rate had a decreasing trend as the 33 treatment fractions progressed. The γ passing rates means±SD of the first fraction was (91.92±3.31)% while the 33th fraction was (85.73±8.75)%. In addition, the standard deviation between the TPS calculations and the EDose measurement results indicated a higher value of the thirty-third treatment for PTVs and organ at risk compared to the first treatment. Conclusions: This study demonstrated that the EDose system is an accurate, efficient method for quality assurance of patient’ radiotherapy plans with remarkable consistency of treatment planning system (TPS).