Blood glucose prediction model construction method based on signal energy characteristics and pulse cycle

The invention relates to a blood glucose prediction model construction method based on signal energy characteristics and pulse cycles, and belongs to the technical field of biomedical signal processing. The method comprises the following steps: S2, preliminary screening: extracting Kaiser-Teager energy characteristics and logarithmic energy entropy characteristics of PPG signal quality, and inputting the Kaiser-Teager energy characteristics and logarithmic energy entropy characteristics into an SVM classifier to divide the signal quality; fine screening: screening and rejecting signals with pulse wave interval periods not conforming to threshold setting; according to the pulse waveform features and the fine screening signal energy features, combining the features of the physiological parameters of the human body to form a 41-dimensional feature vector; obtaining the importance ranking of the features according to xgboost, sequentially adding the features with high importance degrees, and constructing a blood