Automated Skull Thickness Mapping for Transcranial Ultrasound Imaging Systems

Abstract

Skull-induced phase aberrations and signal attenuation limit the performance of transcranial ultrasound imaging, typically confining applications to the temporal region. While adaptive beamforming techniques mostly offer partial correction, they typically lack access to point-specific skull thickness data essential for accurate skull-induced phase aberration compensation. This work presents a skull thickness mapping framework that provides more precise thickness measurements to adjust speed-of-sound (SOS) per probe location, improving phase aberration correction for trans-skull ultrasound applications, including transcranial ultrasound imaging, offering a solution for ultrasound-based transcranial systems. Using 800 CT scans from the MosMed Expanded Brain CT Dataset, cranial bone was segmented via thresholding, and surface meshes were generated through Marching Cubes reconstruction. Skull thickness was estimated by multi-directional ray casting between outer and inner bone boundaries, projected onto 3D surfaces as color-coded thickness maps. Rigid registration using Iterative Closest Point aligned individual skulls to a common anatomical reference, enabling the computation of region-specific mean thickness distributions and variability measures. When implemented in a custom ultrasound platform with a matrix array transducer, the proposed method improved phase aberration correction accuracy compared to fixed-SOS models. By enabling location-specific SOS adjustments, this framework provides a practical and adaptable method for full-skull transcranial ultrasound imaging.

Ähnliche Arbeiten