Denoising embolic doppler signals using modified complex discrete wavelet transform

Stroke is the rapidly developing loss of brain functions, due to a disturbance in the blood vessels supplying blood to the brain. The particles in circulation system, bigger than red blood cells, are accepted as the main reason of emboli. Early and accurate detection of asymptomatic emboli is important in monitoring stroke-prone patients. The main problem in detection of emboli is the identification of embolic signals caused by very small emboli. The amplitude of embolic signals can be very small and advanced signal processing techniques are needed to distinguish these signals from Doppler signals arising from red blood cells. In this study instead of conventional discrete wavelet transform, which is used frequently in processing embolic signals, a modified complex discrete wavelet transform with same computational complexity and better phase shift-invariance property was used for de-noising embolic signals. Results demonstrates that approximately 8 dB improvement is obtained by using modified complex discrete wavelet transform compared to the improvement provided by the conventional discrete wavelet transform, approximately 5 dB.