Breathing adapted radiotherapy using optical real-time thoractic localization

Abstract: Purpose: The aim of this work is to investigate the ability of a laser positioning system for breathing-adapted radiotherapy (BART), also known as respiratory gating. C-RAD installed the Sentinel system at the Sk°ane University Hospital (SUS) in Malm¨o during an appointed time for the purpose of evaluation on the pre-clinical gating prototype. It is also to develop a visual coaching method for the Sentinel system and to investigate the feasibility of this coaching method, by evaluating both reproducibility and stability for the deep-inspiration breath hold (DIBH) technique. Furthermore, to compare two real-time positioning systems for BART; Sentinel vs. Varian Real-time Position Management (RPM) system. Methods and material: Latency measurements were carried out on the Sentinel and the RPM system. The latter system is at present in clinical use at SUS in Malm¨o. A latency measuring device was developed and constructed, using an object of detection attached to a pneumatic piston, an AVR microprocessor, a crystal oscillator, and eight light-emitting diodes. To evaluate the developed visual coaching method, 19 female healthy volonteers were recruited to perform DIBHs using both audio coaching and the developed visual coaching method. The Sentinel system was used to monitor the thoractic AP-PA motion for all volonteers. MATLAB was used to read and plot data from the gating system output files. All reproducibility and stability calculations were also done in MATLAB. Results: All volonteers improved the reproducibility of DIBHs with the developed visual coaching method compared to audio coaching. 38% of the volonteers had improvements less than 1 mm, 46% had an improvement between 1 2 mm, and 15% had an improvement larger than 2 mm. In terms of stability, 23% showed no improvement with the developed visual coaching method. 46% had improvements less than 1 mm, 23% had an improvement between 1 2 mm, and 8% had an improvement larger than 2 mm. For the latency measurements, the arithmetic mean and the standard deviation was calculated to be 192.1 ± 16.5 μs for the RPM system and 94.3 ± 16.0 μs for the Sentinel system. Conclusions: The Sentinel system had a latency comparable with the RPM system. Both stability and reproducibility of DIBHs was improved using the developed visual coaching method. The results indicate that the Sentinel system is a promising system for BART.