Optimization of <scp>MR</scp> ‐Guided Focused Ultrasound System: Comparative Study of Water Bolus and Electrically Optimized Material Using Automated Machine Learning

ABSTRACT Magnetic resonance‐guided focused ultrasound (MRgFUS) is a therapeutic technology designed for the treatment of neurological disorders, enabling precise focal heating under magnetic resonance imaging (MRI) guidance. However, electromagnetic (EM) interaction between the radiofrequency (RF) coil and the MRgFUS system leads to increased specific absorption rate (SAR), reduced RF transmit magnetic (|B 1 + |)‐field homogeneity, and decreased signal‐to‐noise ratio (SNR). In this study, we compared a conventional water bolus containing sodium chloride and sterile water with an electrically optimized material (EOM) optimized using an automated machine learning (Auto‐ML) approach to minimize SAR while maximizing |B 1 + |‐field quality. EM simulation results demonstrated that our EOM achieved significant improvements in |B 1 + |‐field homogeneity and a reduction in peak spatial SAR averaged over 10 g (psSAR 10g ) compared to the conventional water bolus. These findings suggest that Auto‐ML‐based EOM can enhance the safety and efficiency of MRgFUS procedures.