Transcranial ultrafast ultrasound Doppler imaging: A phantom study

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Ultrasound Technology

Background:

• Ultrafast ultrasound Doppler imaging offers high-resolution assessment of cerebral hemodynamics.
• Acoustic impedance mismatch between the skull and soft tissue causes phase aberrations, degrading transcranial imaging quality and biasing blood flow quantification.

Purpose of the Study:

• To develop and validate an aberration correction method for transcranial plane-wave and ultrafast Doppler imaging.
• To improve the accuracy of cerebral blood flow velocity and direction measurements through the skull.

Main Methods:

• A hybrid approach combining deep learning-based skull sound speed modeling with ray theory was employed.
• The method was validated using phantom experiments with a 6.25 MHz linear array and applied to transcranial ultrafast Doppler flow imaging.

Main Results:

• The aberration correction significantly improved imaging quality, reducing locating deviation and improving point scatterer resolution (FWHM).
• Contrast-to-noise ratio (CNR) for circular inclusions increased, and eccentricity decreased, indicating better target delineation.
• Normalized root-mean-square errors for axial velocity in Doppler flow imaging were reduced from 17.67% to 8.02%.

Conclusions:

• The proposed deep learning and ray theory-based method effectively corrects skull-induced phase aberrations in transcranial ultrasound imaging.
• This technique substantially enhances the accuracy and quality of both structural and Doppler imaging of cerebral hemodynamics, paving the way for improved clinical applications.