Medical Ultrasonic Signals and Their Feature Extraction Methods
王威琪,汪源源
DOI: https://doi.org/10.3969/j.issn.1006-6586.2004.01.001
2004-01-01
Abstract:Medical Electronics is a branch of Biomedical Engineering which also involves the electronic fields. The research of signals and information is an important part of Medical Electronics. Usually human signals have strong randomness and background noises. In medical ultrasound imaging technique, with steps of ultrasound transmitting and receiving, the signal detection and analysis, the structure and moving images displaying, new conceptions, methods and techniques of the advanced science and technology are adapted to extract feature parameters which make the clinical diagnosis more accurate, prompt and effective. Here Medical Electronic methods and techniques are mainly introduced to detect, analyse and extract feature for human medical ultrasound signals. The special i nstruments, methods and conceptions concerned in the detection and analysis of the medical signals are presented while their future development is also described. The feature extraction methods of medical ultrasound signals and their effects are introduced respectively in the time domain, frequency or other transform domain, geometry domain and model application according to authors' research work. The Teager energy method for Doppler blood flow signals is an example of time domain method. In the frequency domain, the Short Time Fourier Transform method for the extraction of spectrum parameters is given as an example, while the feature extraction of maximum frequency waveform from carotid arteries using wavelet transform is another example in transform domain. The fractal method used for Doppler blood flow signals analysis is an example in geometry domain. In model methods, an example of zero-pole model parameters extraction for Doppler blood flow signals is given.