Time-frequency and time-scale analysis of embolic signals

Early and accurate detection of emboli is important for monitoring of preventive therapy in stroke-prone patients. Emboli can be detected by using Doppler ultrasound. Most Doppler ultrasound systems employ quadrature demodulation techniques at the detection stage. The windowed Fourier transform has been widely used for the analysis of quadrature signals. The Fourier transform maps directional information in the frequency domain. However, it is not ideally suited to analysis of short duration embolic signals due to an inherent trade-off between time and frequency resolution. Alternatively, the wavelet transform ideally suits to the analysis of nonstationary signals. Mapping directional information in the scale domain is also desirable. A method based on the utilization of complex wavelets and negative scales has been proposed. The results reveal that the wavelet transform provides an optimized time-frequency resolution for analysis and detection of low-intensity embolic signals and the utilization of complex wavelets and negative scales eliminates the intermediate processing stages by mapping directional information in the scale domain.