A fast and practical computation method for magnetic resonance simulators.

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Computational Physics
• Biophysics

Background:

• Magnetic Resonance (MR) simulations are computationally intensive due to the need for high-resolution temporal updates of isochromat magnetizations.
• Traditional methods often spend significant time processing repetitive radiofrequency (RF) pulse subsequences.

Purpose of the Study:

• To develop a computationally efficient and practical method for MR simulations.
• To address the high computational cost associated with simulating complex RF pulse sequences.

Main Methods:

• Proposes a novel method that computes and caches combined transitions for RF pulse subsequences.
• Replaces iterative updates with a single computation of combined transitions for repeated sequences.
• Employs a least recently used (LRU) algorithm for managing cached combined transitions.

Main Results:

• Achieved a simulation acceleration of approximately 20 times compared to conventional methods.
• Successfully simulated 3.9 million isochromats using spin-echo sequences.
• Demonstrated the feasibility of performing complex MR simulations in approximately 3.5 minutes on a laptop.

Conclusions:

• Introduces an efficient MR simulation technique by optimizing the processing of RF pulse sequences.
• The proposed method enhances the practicality of MR simulations, enabling their execution on diverse computing hardware, including laptops.