Multi-echo MR thermometry in the upper leg at 7 T using near-harmonic 2D reconstruction for initialization

Area of Science:

• Medical Imaging
• Biophysics
• Electrical Engineering

Background:

• Magnetic Resonance (MR) imaging utilizes radiofrequency (RF) waves, necessitating monitoring of tissue heating to ensure safety.
• Current safety protocols rely on predefined thresholds for specific absorption rate (SAR), but advanced validation methods are needed.
• MR thermometry offers a potential solution for assessing RF-induced heating and validating thermal simulations.

Purpose of the Study:

• To develop a precise and accurate in vivo MR thermometry method for validating thermal simulations.
• To enhance the safety assessment of radiofrequency (RF) exposure during MR imaging.
• To provide a tool for investigating local peak temperatures and informing temperature-based safety guidelines.

Main Methods:

• A harmonic initialized model-based multi-echo approach was developed, integrating existing water/fat separation and 2D reconstruction techniques.
• The method was tested in vivo on the human thigh using a 7 Tesla (7T) multi-transmit array across multiple RF shim settings and volunteers.
• Data acquisition and analysis focused on precise temperature rise measurements.

Main Results:

• The proposed MR thermometry method achieved significantly improved precision (0.09°C) compared to previous fat-referenced techniques (0.19°C).
• Measured temperature rise distributions showed good relative agreement with subject-specific thermal simulations for various RF shim settings.
• The results demonstrate enhanced accuracy and reliability of the developed thermometry approach.

Conclusions:

• The developed MR thermometry method offers high precision, showing significant potential for validating RF safety simulations.
• This technique can be applied to various RF safety applications, improving the understanding of tissue heating during MRI.
• The findings support the use of advanced MR thermometry for robust safety assessments in MR imaging.