Adaptive quadrature demodulation for ultrasound tissue harmonic imaging

Frequency dependent attenuation of ultrasound causes the center frequencyof received echoes to be downshifted along depth. This is particularly severe ona harmonic component rather than a fundamental one because higher ultrasoundfrequency experiences higher attenuation. The different degrees of downshiftingalong with ultrasound frequency should be considered when quadraturedemodulation (QDM) is carried out to secure the best signal-to-noise ratio(SNR). In the conventional second harmonic imaging, however, a constantattenuation coefficient is typically assumed for estimating the center frequencyof the harmonic components in QDM. This assumption may not be valid becausethere is the attenuation variation in an imaging area. To obtain the best SNR,therefore, this paper proposes an automatic center frequency estimator based onautoregressive (AR) modeling for the ultrasound second-harmonic tissue imaging.In the proposed method, the estimated center frequency of a second-harmoniccomponent is directly used for dynamic QDM and dynamic low-pass filtering (LPF).The performance of the proposed method was evaluated by Field II simulation andin-vivo experiments. The experimental results demonstrated that the proposedmethod is capable of providing higher contrast resolution while reducinghigh-frequency noise under visual examination due to the improvement of SNR.