Predicting long-term temperature increase for time-dependent SAR levels with a single short-term temperature...

Area of Science:

• Biomedical Engineering
• Medical Imaging Physics

Background:

• Accurate prediction of radiofrequency-induced heating is crucial for Magnetic Resonance Imaging (MRI) safety.
• Existing methods for temperature prediction can be computationally intensive, limiting real-time applications.

Purpose of the Study:

• To present a novel, rapid method for predicting in vivo temperature changes during MRI.
• To assess the accuracy and computational efficiency of this new method compared to traditional simulations.

Main Methods:

• Characterization of the temperature response using a linear heat equation and impulse-response (IR) concepts.
• Initial characterization and prediction using Pennes' bioheat equation, incorporating core body temperature changes and local perfusion.
• Introduction of time-varying corrections for local perfusion to refine predictions.

Main Results:

• The IR-based method achieved maximum temperature predictions within 1% accuracy of full finite difference simulations.
• The novel method required less than 3.5% of the computation time compared to traditional methods.
• Corrections for temperature-dependent perfusion were computationally negligible and adjustable for conservatism.

Conclusions:

• A rapid and accurate method for predicting in vivo temperature increase during MRI examinations is feasible.
• This approach allows for efficient safety monitoring by predicting temperature changes related to specific energy absorption rate (SAR).