Skip to main content

Table 1 Sensitivity test cases and modified parameters for introducing errors into prediction model

From: Investigation of a real-time EPID-based patient dose monitoring safety system using site-specific control limits

Error Class Simulated Errors Plan modification methods
Patient position misalignment Patient setup errors = 5, 7, 10 mm Horizontal spatial shift of CT images data set, then recalculate predicted EPID data; only gantry angle at zero is tested.
MU errors Increasing 5, 7, 10 % Decrease MU in treatment plan then calculate predicted EPID data; this simulates overdose delivery to patient.
Wrong Patient or Plan Incorrect patient same treatment site Apply different patient CT data for predicted EPID data. This simulates wrong patient being treated.
Wrong gantry angle Correct plan (patient) incorrect field or wrong gantry angle Re-ordered gantry angle in treatment plan (exchanged gantry angle between treatment fields) then re-calculated predicted EPID data. This simulates wrong gantry angle in IMRT.