Encrypted login | home

Program Information

Generalized Beam Modulation Based Dosimetric Leaf Gap Determination Scheme

T Chen

T Chen1*, N Yue2 , (1) Rutgers University, New Brunswick, NJ, (2) Rutgers University, New Brunswick, NJ


SU-I-GPD-T-388 (Sunday, July 30, 2017) 3:00 PM - 6:00 PM Room: Exhibit Hall

Purpose: To establish a generalized scheme to determine the appropriate Dosimetric Leaf Gap (DLP) from the ion chamber measurement of point dose in the verification plan of RapidArc plan for real patients.

Methods: We first performed Varian standard procedure to determine the initial value of DLP of HD120 MLC on a newly installed TrueBeam STx. The DLP value has been put into the Varian Eclipse TPS for initial beam configuration. We retrospectively calculated the point dose in the solid water phantom based verification plan for 41 patients who received 6X RapidArc. The dose actually delivered to the point was measured using A1SL ion chamber. The dose difference between the measurement and the plan as well as the correlation between the dose difference and the plan modulation has been fed back to a generalized optimization equation to minimize the correlation between the plan modulation and the dose difference. Updated value of DLG was used to recalculate the verification plans iteratively until a minimization has been achieved. The final DLG value was used for clinical.

Results: For 41 6X RapidArc patients, verification plans with the initial DLG of 0.35mm were delivered to a TomoTherapy cheese phantom. The point dose difference between ion chamber measurements and the verification plans showed correlation (corr = -0.597, p<0.001) to the plan modulation factor. After iterative optimization using 0.3mm as the step size of DLG, it was determined that the optimal DLG value was 1.25mm. By using 1.25mm as the DLG value, the RSM dose difference between ion chamber measurements and the verification plans reduced from 1.62% to 0.71% for all patients. The correlation coefficient between the dose difference and the plan modulation reduced to 0.1187 (p=0.2354).

Conclusion: We developed a clinical data driven formalism to automatically determine the optimal value for DLG for 6X RapidArc plans.

Contact Email: