Room: AAPM ePoster Library
Purpose: Present an in vivo dosimeter with real-time positional information for error detection in brachytherapy.
Methods: A plastic scintillation dosimeter prototype was coupled with an 5DOF electromagnetic (EM) sensor read by the Aurora V3 system (NDI, Canada). The scintillator used was a 3 mm long BCF-60 of 0.5 mm diameter (Saint-Gobain Crystals). The EM sensor was 4 mm long and had a 0.4 mm diameter; it was placed at the dosimeter’s tip with a distance of 2cm from the scintillator. Position sampling rate was 40/s while the scintillator was sampled at 100 000/s using photomultiplier tube from Hamamatsu (series H10722) connected to a data acquisition board (NI-DAQ board , National Instruments, USA). Furthermore, dose rate was converted to dosimeter-source distance and the position of the dosimeter deconvolved using EM. Dwell time was calculated with the dose rate by a gradient search. All measurements were performed with an afterloader unit (Flexitron-Elekta AB, Sweden) to determine both positioning and dwell time precision. Indexer length errors between 1 to 15 mm were simulated by moving the source. The dose difference measured was converted to a shift distance and was compared to the simulated errors.
Results: The mean absolute errors for time computation is 0.3±0.1s. Dose rate measurement lies within 3.6 % from TG-43 up to a distance of 10 cm from the source. Dose measurement can detect all indexer’s errors higher than 3 mm at distances under 4 cm, and down to 1 mm below 2 cm from the source. These also corresponds to catheter shift detection rates along the insertion direction.
Conclusion: This work demonstrates that integrating an EM tracking sensor to an energy independent plastic scintillation dosimeter can be used to detect all indexer’s length or catheter shift errors by removing the ambiguity on the detector positioning in the applicator/catheter space.