A spatiotemporal coherence-based approach for optimizing transcranial ultrasound imaging sequences

Abstract

Functional ultrasound imaging (fUSI) is a promising new neuroimaging modality for preclinical and niche clinical applications, but translation to noninvasive use in adults remains unresolved due to skull-induced image degradation. Attenuation leads to low signal-to-noise ratio (SNR), whereas clutter leads to low signal-to-clutter ratio (SCR). Coded excitation or microbubbles could increase SNR, whereas adaptive beamforming, phase aberration correction, or harmonic imaging could increase SCR. However, it’s unknown how the relative importance of noise and clutter varies across demographics. To optimize a demographic-specific transcranial imaging sequence, a tool to measure SNR and SCR and assess the impact of sequence modifications is needed. We propose using a spatiotemporal coherence technique [DOI:10.1109/TUFFC.2022.3152717] to measure SNR and SCR with coded excitation [DOI:10.1109/IUS46767.2020.9251650] for different parameters like code length and voltage. We conducted a transcranial imaging study in five volunteers (median age 25, Caucasian, three males, two females) and used focused M-Mode data to measure SNR and SCR. With optimally chosen parameters, we improved SNR by 16.87 ± 2.18 dB and 13.18 ± 1.14 dB with 65 bit and 25 bit codes without introducing additional clutter (SCR gains were −0.29 ± 0.67 dB and 0.13 ± 0.84 dB with 65 and 25 bit codes). These improvements will enable clinical translation of transcranial fUSI.