Elementwise approach for simulating transcranial MRI-guided focused ultrasound thermal ablation

Area of Science:

• Neurosurgery
• Medical Imaging
• Acoustic Modeling

Background:

• Transcranial MRI-guided focused ultrasound (TcMRgFUS) offers a noninvasive approach for neurosurgery.
• Accurate modeling of thermal ablation is crucial for effective and safe treatment delivery.
• Existing modeling methods may face challenges in computational speed and accuracy for complex acoustic fields.

Purpose of the Study:

• To develop and validate an elementwise computational approach for modeling TcMRgFUS thermal ablation.
• To enable rapid calculation of pressure fields and temperature rise for beam steering and aberration correction.
• To optimize the relationship between CT-derived skull density and acoustic properties for improved prediction accuracy.

Main Methods:

• Simulated individual elements of the phased array transducer for rapid pressure field calculation (~2.5 s).
• Modeled 431 sonications across 32 patients, applying treatment-specific phase and magnitude values.
• Systematically varied CT-derived skull density to acoustic property relationships, optimizing against MR temperature imaging (MRTI) data.

Main Results:

• Optimized model showed good agreement with MRTI (R²: 0.74 for patients 1-32; 0.71 for patients 33-72).
• Predicted heating dimensions and obliquity correlated well with MRTI (R²: 0.62, 0.74, respectively).
• Observed significant variation in energy requirements for ablation (3.3-36.1 kJ) and more spatially diffuse measured heating.

Conclusions:

• The elementwise approach enables rapid, parallelizable computation of 3D heating patterns in TcMRgFUS.
• Optimization of skull density-acoustic property relationships enhances prediction accuracy.
• This method shows promise for modeling TcMRgFUS, requiring further validation in larger patient cohorts.