Wavelet Analysis of Non-stationary Signals in Medical Cyber-Physical Systems (MCPS)

摘要

The advantages of multichannel medical cyber-physical systems (MCPS), which are designed to receive and process signals of human biological rhythms (EEG, ECG, blood pressure) at the remote server and to issue diagnostic conclusions, are discussed. The paper presents new data processing algorithms for MCPS based on continuous wavelet transform (CWT). The proposed method provides an array of parameters characterizing frequency restructuring of medical signals in different frequency bands. Since frequency fluctuations in brain and heart rhythms are closely related to different processes in human organism, the obtained data can allow us: to identify the disease at its early stages; to test the adaptive capacity of the human organism; to give diagnostic reports on cardiovascular and nervous systems; to analyze changes in rhythm during biofeedback sessions. The techniques set forth in the paper can help in the creation of a unique "rhythmic portrait" of a person to diagnose his physiological state.