Systematic quality assurance of the BSD2000-3D MR-compatible hyperthermia applicator performance using MR temperature imaging

Background: Radiofrequency (RF) mild hyperthermia (40 degrees C-44 degrees C for 60 minutes) is an effective adjuvant treatment for several types of cancer. To ensure treatment efficacy, quality assurance (QA) is necessary. This study presents the first systematic 3D characterisation of the heating performance of the commonly used Pyrexar BSD2000-3D MR-compatible hyperthermia applicator using magnetic resonance temperature imaging (MRTI). Methods: A reproducibly positioned phantom was heated with a power of 1000 watts during the 12.4 min needed to measure eight temperature distributions using MRTI. The target heating location was systematically varied between experiments. We analysed focus shape characteristics, steering accuracy, focus deformation due to steering, presence of off-target heating and reproducibility. Results: The mean maximum temperature increase was 5.9 +/- 0.4 degrees C. The mean full width half maximum (FWHM) was 14.4 +/- 0.5 cm in the XY plane and 24.5 +/- 0.8 cm in Z-direction. The mean steering error was 0.4 +/- 0.2 cm. The focus shape slightly varied between experiments, depending on steering distance in Y-direction. Off-target heating was not detected. Reproducibility of the focus amplitude and shape was determined by comparing the mean deviation from the mean temperature in the central slice was 0.3 +/- 0.2 degrees C. Conclusion: The Pyrexar BSD2000-3D MR-compatible applicator provides robust and reproducible heating. The upper boundary of the 95% confidence interval of the spatial steering accuracy is 0.9 cm, i.e. sufficient to fulfil the criterion of <= 0.2 degrees C temperature variation due to positioning errors as defined by Canters et al.