Vergleich nichtinvasiver MRT-Verfahren zur Temperaturmessung für den Einsatz bei medizinischen Thermotherapien

Abstract

Novel methods for hyperthermia tumor therapy, such as high-intensity focused ultrasound (HIFU) or laser-induced thermotherapy (LITT), require accurate non-invasive temperature monitoring. Non-invasive temperature measurement using magnetic resonance imaging (MRI) is based on the analysis of changes in longitudinal relaxation time (T1), diffusion coefficient (D), or water proton resonance frequency (PRF). The purpose of this study was the development and comparative analysis of the three different approaches of MRI temperature monitoring (T1, D, and PRF). Measurements in phantoms (e.g., ultrasound gel) resulted in the following percent changes: T1-relaxation time: 1.98% /°C; diffusion coefficient: 2.22% /°C; and PRF: −0,0101 ppm /°C. All measurements were in good agreement with the literature. Temperature resolutions could also be measured from the inverse correlation of the data over the whole calibration range: T1: 2.1 ± 0.6 °C; D: 0.93 ± 0.2 °C; and PRF: 1.4 ± 0.3 °C. The diffusion and PRF methods were not applicable in fatty tissue. The use of the diffusion method was restricted due to prolonged echo time and anisotropic diffusion in tissue. Initial tests with rabbit muscle tissue in vivo indicated that MR thermometry via T1 and PRF procedures is feasible to monitor the local heating process induced by HIFU. The ultrasound applicators in the MR scanner did not substantially interfere with image quality.