Spectral analysis of episodic rhythmic variations in the cutaneous electrogastrogram

The cutaneous recording of gastric electrical activity is called the electrogastrogram (EGG). Gastric electrical dysrhythmic events associated with abnormal conditions of the stomach may be detected from the EGG. An adaptive spectral analysis method based on autoregressive moving average (ARMA) modeling has previously been developed. The ability of the previous method to detect gastric dysrhythmic events from the EGG is studied. A series of bench tests simulating typical problems with the analysis of nonstationary electric potentials was conducted. The application of the adaptive spectral analysis method to dysrhythmic events and rhythmic variations of the gastric slow wave is presented. The adaptive spectral analysis approach provides several advantages: narrow frequency peaks permitting more precise frequency identification, determination of changes in frequency components at any time point, and enhanced interpretation of cutaneous EGG recordings.