Nonlinearity parameter imaging of local estimates using spatial compounding

Avilés E.; Lavarello R.; Coila A. · IEEE Ultrasonics, Ferroelectrics, and Frequency Control Joint Symposium, UFFC-JS 2024 - Proceedings (2024) · DOI: 10.1109/uffc-js60046.2024.10793740

Resumen

Nonalcoholic fatty liver disease (NAFLD) is a major global health concern, with current diagnostic methods like ultrasound being limited in specificity, particularly in early stages. The nonlinearity parameter (B/A) is increasingly recognized as a quantitative ultrasound tool for characterizing tissue, given that fatty liver tissue exhibits higher B/A values compared to healthy tissue. However, existing methods for estimating B/A local maps often have limited precision. To address these limitations, a new approach is proposed using spatial compounding (SC) to improve local B/A estimation accuracy of the depletion method. The proposed method aligns B/A cumulative maps of the depletion method through rotation to better satisfy inversion assumptions for local B/A estimation, thereby enabling SC to reduce noise and artifacts. Numerical phantom simulations with B/A of 11 and 6 for inclusion and background, respectively, were used to evaluate this method with and without Total Variation (TV) regularization. Results suggest that SC decreases the variance in B/A local map estimation compared to using a single transmitted angle without steering. Additionally, reducing the number of steering angles can increase the frame rate of the acquisition while maintaining image quality during the compounding process. The results suggest potential for enhanced tissue characterization of B/A estimates by improving the accuracy using spatial compounding.

Instituciones: Pontificia Universidad Católica del Perú