Rapid Quantification of CEST Parameters Based on Multi-Starting Point Optimization and Bloch-Mcconnell Analytical Solution

Abstract

This study presents an innovative approach that combines multi-start (MS) optimization with the Bloch-McConnell eigenspace analytical solution for the rapid and accurate quantification of chemical exchange saturation transfer parameters. The core advantage of this method lies in the global search capability of the MS algorithm, which reduces the reliance on a single initial value and mitigates the risk of getting trapped in local optima, a common issue with traditional fitting methods. Additionally, by utilizing the exchange-dependent relaxation rate in the rotating frame, the computational process is simplified, significantly enhancing the stability and reliability of parameter estimation. Simulation results show that, compared to traditional optimization methods such as genetic algorithms, particle swarm optimization, and global search algorithms, our method reduces computation time from 9358.9 seconds to 2175.1 seconds, achieving a 4.3-fold acceleration while maintaining lower reconstruction errors. A 7 T magnetic resonance imaging (MRI) rat brain experiment validated the robustness of this method under different B1 field conditions, and the generated high-resolution parameter maps accurately distinguish tumor from normal tissue, demonstrating the potential of this method for real-time chemical exchange saturation transfer (CEST) imaging. By reducing computation time and improving accuracy, this method provides an efficient and reliable tool for clinical and research applications.