In vivo optimisation study for multi-baseline MR-based thermometry in the context of hyperthermia using MR-guided high intensity focused ultrasound for head and neck applications

Abstract

Purpose: In this in vivo study, the feasibility to perform hyperthermia treatments in the head and neck using magnetic resonance image-guided high intensity focused ultrasound (MRgHIFU) was established using a porcine acute model. Materials and methods: Porcine specimens with a weight between 17 and 18kg were treated in the omohyoid muscle in the neck. Hyperthermia was applied with a target temperature of 41°C for 30min using a Sonalleve MRgHIFU system. MR-based thermometry was calculated using water-proton resonance frequency shift and multi-baseline look-up tables indexed by peak-to-peak displacement (Dpp) measurements using a pencil-beam navigator. Three hyperthermia experiments were conducted at different Dpp values of 0.2, 1.0 and 3.0mm. An optimisation study was carried out to establish the optimal parameters controlling the multi-baseline method that ensured a minimisation of spatial-average peak-to-peak temperature (TSA-pp) and temperature direct current bias (TSA-DC). Results: The multi-baseline technique reduced considerably the noise on both TSA-pp and TSA-DC. The reduction of noise was more important when Dpp was higher. For Dpp=3mm the average (±standard deviation (SD)) of TSA-pp and TSA-DC was reduced from 4.5 (± 2.5) and 2.5 (±0.6)°C, respectively, to 0.8 (± 0.7) and 0.09 (± 0.2)°C. Conclusions: This in vivo study showed the level of noise in PRFS-based thermometry introduced by respiratory motion in the context of MRgHIFU hyperthermia treatment for head and neck and the feasibility of reducing this noise using a multi-baseline technique.