A forward-backward subsequence smoothing based eigen approach to clutter rejection in color flow imaging

In color flow image (CFI), the prevalent clutter filters fail to accurately separate blood flow signal from the demodulated ultrasound echo signal containing undesired clutter with single-ensemble (S-E) samples of small pulse size. A novel clutter filter design framework based on forward-backward subsequence smoothing (FBSS) as an extension to the prevalent eigen-based method is proposed. The proposed FBSS algorithm simulates the multi-ensemble behaviors with only single-ensemble (S-E) samples such that a fixed-length segment of each S-E sample is considered as a short conventional S-E vector. By moving forward and backward one sample, many vectors of the same-length can be created. This process can be repeated until all sample points in the original S-E sequence is exhausted and the collection of these vectors are considered as a simulated M-E ensemble. The pulse size needed to guarantees a nonsingular property of autocorrelation matrix in the FBSS framework is usually three-quarter of those used in the Hankel framework. As a consequence, FBSS leads to a good discrimination between the clutter and blood flow components for the low-velocity blood flow signals. Based on the results obtained from a series of simulations, the proposed FBSS eigen-based filter is shown to achieve a superior performance to the state-of-the-art regression and Hankel-SVD filters in estimating the blood flow velocities.