An Application of Wavelet Transform (WT) and Independent Component Analysis (ICA) for Electrogastrographic (EGG) Signals Artifacts Detection
暂无分享,去创建一个
Electrogarstrogram (EGG) is an electric signal that is propagated through the muscles of stomach controlling muscles contractions and measuring stomach nerve activity before and after food ingestion. It is easy to perform, noninvasive and relatively inexpensive test which therefore has become an attractive method for physiologic and pathophysiologic examinations of the stomach. The main component of gastric myoelectrical activity, called gastric slow waves, has a frequency about 3 cycles/min (0.05 Hz). As a result the EGG signal requires longer recording time (usually more than 1 hour). Furthermore EGG is a weaker signal then other bioelectric signals, such as ECG or respiratory noise, so the EGG is usually contaminated by artifacts, which damage the recorded data and make analysis very difficult or impossible. In order to use EGG as diagnostic tool the artifacts have to be first detected and then automatically eliminated before the analysis starts.
[1] Aapo Hyvärinen,et al. Fast and robust fixed-point algorithms for independent component analysis , 1999, IEEE Trans. Neural Networks.
[2] Aapo Hyvärinen,et al. The Fixed-Point Algorithm and Maximum Likelihood Estimation for Independent Component Analysis , 1999, Neural Processing Letters.
[3] E.J. Tkacz,et al. Supervised and Unsupervised Learning Systems as a Part of Hybrid Structures Applied in EGG Signals Classifiers , 2005, 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference.