Initial validation of a diaphragm tracking system for multiple breath-hold volumetric modulated arc therapy of abdominal tumors: A phantom study

BackgroundThe breath-hold radiotherapy has been increasingly used to mitigate interfractional and intrafractional breathing impact on treatment planning and beam delivery. Previous techniques include body surface measurements or radiopaque metal markers, each having known disadvantages.PurposeWe recently proposed a new markerless technique without the disadvantages, where diaphragm was registered between DRR and fluoroscopic x-ray projection images every 180 ms during VMAT delivery. An initial validation of the proposed diaphragm tracking system (DiaTrak) was performed using a chest phantom to evaluate its characteristics.MethodsDiaphragm registration was performed between DRR and projection streaming kV x-ray images of a chest phantom during VMAT delivery. Streaming data including the projection images and the beam angles were transferred from a linac system to an external PC, where the diaphragm registration accuracy and beam-off latency were measured based on image cross correlation between the DRR and the projection images every 180 ms.ResultsIt was shown that the average of the beam-off latency was 249.5 ms and the average of the diaphragm registration error was 0.84 mmConclusionsInitial validation of the proposed DiaTrak system for multiple breath-hold VMAT of abdominal tumors has been successfully completed with a chest phantom. The resulting beam-off latency and the diaphragm registration error were regarded clinically acceptable.