The invention discloses a multi-lead electrocardiogram (ECG)
signal composite
feature extraction method and a corresponding
monitoring system, and relates to the field of
ECG signal analysis and detection. The method comprises the following steps: 1, extracting statistical features or morphological statistical features and
wavelet energy entropy features of the morphology of single-lead ECG signals; 2, repeating the step 1 to acquire and fuse statistical features or the morphological statistical features and the
wavelet energy entropy features of the lead morphology; the
system comprises a
feature extraction module; the
feature extraction module comprises a dynamic link
library, a feature extraction unit and a
feature fusion unit, and is used for extracting the statistical features or themorphological statistical features and the
wavelet energy entropy features of the morphology. The method is used for extracting the statistical features or the morphological statistical features and the wavelet energy entropy features of the morphology of the ECG signals, fully characterizes the local features of the signals, enhances the feature expression ability, and achieves the effects of accurately capturing the small short dynamic changes of the ECG signals and the morphological changes of the complex
ECG waveforms as well as accurately identifying the
normal state and abnormal state ofthe ECG signals.