CARDIAC DOPPLER BLOOD-FLOW SIGNAL ANALYSIS .2. TIME-FREQUENCY REPRESENTATION BASED ON AUTOREGRESSIVE MODELING

Doppler spectrograms obtained by using autoregressive (AR) modelling b ased on the Yule-Walker equations were investigated, A complex AR mode l using the in-phase and the quadrature components of the Doppler sign al was used to provide blood-flow directions. The effect of model orde rs on the spectrogram estimation was studied using cardiac Doppler blo od flow signals taken from 20 patients. The 'final prediction error' ( FPE) and the Akaike's information criterion' (AIC) provided almost ide ntical results in model-order selection. An index, the spectral envelo pe area (SEA), was used to evaluate the effect of window duration and sampling frequency on AR Doppler spectrogram estimation. The statistic al analysis revealed that the SEA obtained from AR modelling was not s ensitive to window duration and sampling frequency. This result verifi ed the consistency of the AR Doppler spectrogram. The white-noise char acteristics of the AR modelling error signal indicated that the Dopple r blood-flow signal can be adequately modelled as a complex AR process . With appropriate model orders, AR modelling provided better Doppler spectrogram estimates than the periodogram.