Click here to


Are you sure ?

Yes, do it No, cancel

Initial Investigation of the SX1 Dual-Layer Multileaf Collimator Tongue-And-Groove Effect

TY Lim*, E Flores-Martinez , X Fave , S AM Lloyd , I Dragojevic , D Hoffman , T Atwood , K L Moore , G Kim , UC San Diego, La Jolla, CA


(Tuesday, 7/31/2018) 10:30 AM - 11:00 AM

Room: Exhibit Hall | Forum 4

Purpose: To characterize the Varian Halcyon SX1 dual-layer multileaf collimator (MLC) system tongue-and-groove effect.

Methods: We performed relative and absolute dose measurements for static, IMRT, and VMAT plans. We delivered an alternating leaf pattern to examine the relative tongue-and-groove effect dependence on MLC banks. We also delivered this pattern at the 4 cardinal gantry angles to investigate the relative tongue-and-groove effect dependence on gantry angles. To investigate the absolute dose profile discrepancy between measurements and modelling of the tongue-and-groove effect, we delivered abutting static fields and compared the film measurement to the dose distribution predicted by the treatment planning system (TPS; Eclipse AAA15.1.51). We also evaluated the chamber-based patient-specific quality assurance results of the first 100 IMRT patient plans treated at our clinic to understand the potential clinical impact of the tongue-and-groove effect modelling. To scrutinize the tongue-and-groove effect in VMAT arcs, we applied the same chamber measurements for 2 simple test plans with and without leaf travel across the chamber measurement point, and performed chamber measurements for 10 VMAT test plans in 5 disease sites (prostate, brain, head-and-neck, gynecology, and spine).

Results: No bank or gantry angle dependence for the tongue-and-groove effect was observed. Comparison of the abutting static field measurements to the TPS prediction showed that the TPS underestimated the under-dosage between adjacent leaf pairs. The amount of deviation depends on the inspected location within the field due to the flattening-filter-free beam. Comparing chamber measurements to TPS-predicted dose in IMRT and VMAT plans, we observed a trend towards enhanced underestimation in highly-modulated plans. Specifically, in the 10-patient test cohort, the dose deviation ranged from -5.45% to -2.43% measured with the PinPoint chamber, and ranged from -4.47% to 0.36% with the Farmer chamber.

Conclusion: The SX1 MLC system tongue-and-groove effect has been comprehensively characterized.


MLC, Intensity Modulation, Linear Accelerator


TH- External beam- photons: Quality Assurance - Linear accelerator

Contact Email