Electrogastrogram (EGG) is the cutaneous recording of gastric electrical activity. Some patients with nausea, vomiting, early satiety, and patients with long term insulin-dependent diabetes exhibit gastric electrical rhythm disturbances. We are interested in investigating the EGG to determine if abnormalities of the EGG occur in association with those gastric complaints. The aim of this document is to investigate long term diabetics with a history of such gastric dysrhythmia complaints in an effort to further characterize the frequency of abnormalities of the EGG in real clinical diabetic gastroparesis patients. For feature extraction purposes, we have performed Wavelet-Packet analysis of EGG signals and have obtained power spectral density of wavelet-packet details and approximation coefficients in diabetic patients suffering from gastric dysrhythmia (tachygastria, bradygastria) and healthy volunteers. Wavelet-Packet analysis provides details of changes in frequency components at any time point, and enhanced interpretation of cutaneous EGG recordings. In conclusion, this paper has shown that the Wavelet-Packet analysis method is very useful in the analysis of cutaneous EGG recordings, especially in detecting dysrhythmic events.