Author: Niko Ahjoniemi, Antti Kulmala, Katja Peltola 👨🔬
Affiliation: Clinical Research Institute HUS, Helsinki University Hospital 🌍
Purpose: To determine the backlash of a multileaf collimator (MLC) using four measurement methods: motor encoder (primary readout), linear potentiometer (secondary readout), amorphous silicon panel (portal image) and IR sensor (built-in calibrator).
Methods: To record the actual movement of the MLC leaves, two dynamic fields were created: one for the closing and another for the opening motions of the leaves. Leaves were moving at maximum speed between their extreme positions, first with all leaves and then with movement in which either even or odd numbered moved at once. For each leaf, average backlash was determined using the actual leaf positions measured by motor encoder, linear potentiometer, and amorphous silicon panel, and then the results are compared to the result of the vendor’s built-in test. Each field was delivered in collimator angle 90° and 0°, gantry angle being 90°. Three ‘Varian Standard 120M’/TrueBeam (Varian Medical Systems Inc., Palo Alto) MLCs were studied.
Results: The backlash determined by the amorphous silicon panel and the built-in backlash test show the most similar results. The coefficient of determination (R2) was found to be 0.48 in collimator angle of 90° and 0.23 at a collimator angle of 0°, indicating a moderate relationship. A weak correlation was found between backlash determined by the linear potentiometer and the built-in test, R2 = 0.19. Furthermore, the relationship between the backlash determined by the motor encoder and the built-in test was weak (R2 = 0.23 at collimator angle 90°, R2 = 0.21 at collimator angle 0°).
Conclusion: Backlash determined by amorphous silicon panel at a collimator angle 90° is found to align best with the vendor’s built-in test, compared to results based on motor encoder and the linear potentiometer. Furthermore, collimator angle 90° reveals stronger relationships between the measured backlash values than collimator angle 0°.