Hilbert-Huang Transform Based Doppler Blood Flow Signals Analysis

In this paper a feature extraction method based on Hilbert-Huang transform (HHT) is proposed in order to investigate the relationship between Doppler ultrasound blood flow signal (DUBFS) of wrist radial artery and the pathological changes of internal organs of human body. Two kinds of empirical model decomposition (EMD) based de-noising methods were compared in data preprocessing progress, and finally EMD interval thresholding (EMD-IT) method was selected to remove the noise of blood flow signal. Then the extracted features based on HHT were applied to classify healthy people from cholecystitis patients and nephritis patients, respectively. Experiment results showed that HHT was an effective method for blood flow signal analysis. Also, the extracted features had promising discriminating ability between the healthy people and different kinds of patients.