Adaptive Clutter Suppression for Ultrafast Ultrasound Imaging using Principal Component Analysis and Fuzzy C-Means Clustering

Inhalt:
Atherosclerosis is a major risk factor for cardiovascular diseases, and its progression is closely associated with the blood flow velocity profile (BFVP). Ultrafast ultrasound imaging based on coherent plane-wave compounding (CPWC) improves spatial resolution but remains susceptible to motion artifacts and clutter interference in complex flows. To address these limitations, we propose a clutter suppression method based on principal component analysis (PCA) with adaptive thresholding via fuzzy C-means (FCM) clustering. The beamformed RF signals are decomposed into orthogonal principal components, and blood flow components are adaptively identified by clustering the associated eigenvalues. The corresponding blood flow components are subsequently reconstructed by retaining the principal components identified as blood flow via FCM clustering, thereby suppressing tissue clutter and noise. The performance has been evaluated in computer simulation for a model of the carotid artery. Results show that the proposed method achieves significantly lower normalized root mean square error (NRMSE) in estimating blood flow velocity profiles, reducing NRMSE by 30.4%–39.0% across varying flow rates compared to conventional SVD-based filtering. The proposed method demonstrates enhanced robustness at higher flow velocities, offering a promising tool for improving the measurement accuracy of BFVP and early cardiovascular diagnosis.