| Summary: | 碩士 === 逢甲大學 === 自動控制工程所 === 97 === On the contrary, the dynamic mean used in the nonlinear and non-stationary signals is other than the usually average mean obtained from a stationary period. Biomedical signals such as blood pressure and ECG, etc. are typical non-stationary signals. It can’t be treated as average mean of the stationary calculation. The dynamic mean may be defined as the average of the maximum and minimum at instant time. It could be found by using Empirical Mode Decomposition(EMD) developed by Norden E. Huang. However, it isn’t eligible to compute the dynamic mean in real time.
The purpose of this paper is to propose a technique that combines EMD, pattern recognition, estimation and prediction theory, estimating the real time dynamic means effectively. From ten healthy subjects(20~28 years old) the simulation results reveal that the mean errors were less than 1 mmHg and indicates that the on-line estimation of dynamic mean blood pressure is feasible.
|
|---|
