This thesis proposes efficient clutter filtering methods to increase the detectable Doppler frequency range in pulse inversion color flow imaging. The proposed π-initialization IIR filter and band selective regression filter can reject more effective...
This thesis proposes efficient clutter filtering methods to increase the detectable Doppler frequency range in pulse inversion color flow imaging. The proposed π-initialization IIR filter and band selective regression filter can reject more effectively the clutter in a wider frequency range than the conventional filters.
The proposed π-initialization technique minimizes the transient response of an IIR filter to remove the clutter signal centered at half the pulse repetition frequency (PRF) due to the pulse inversion scheme. The band selective regression filter can suppress the clutter signal by more than 100dB by using a base function composed of frequency components in the clutter frequency band.
Computer simulation results show that the π-initialization IIR filter provides almost the same performance as a projection initialization IIR filter, with a reduced computational complexity. It was also verified that the band selective regression filter can increase the Doppler detection frequency range by 0.05 PRF to 0.1 PRF and improve the accuracy of the estimated blood flow velocity by up to 8%, with only a small increase in hardware complexity.