Recognition method and device of somatosensory evoked potential components based on support vector machine

Abstract

The invention discloses a support vector machine-based recognition method and device for somatosensory evoked potential components, including an acquisition and construction unit for acquiring median nerve somatosensory evoked potentials from known sources, and performing time-frequency conversion to obtain median nerve somatosensory evoked potential time-frequency components Figure, the time-frequency component features of the median nerve somatosensory evoked potential time-frequency component diagram representing the functional state of the cervical spine are matrix transformed to obtain the time-frequency component feature vector, and the time-frequency component feature vector is used as input to construct a multi-stage support vector classifier The identification unit is used to classify and identify the time-frequency component of the median nerve somatosensory evoked potential to be identified according to the multi-stage support vector classifier, and determine the source corresponding to the time-frequency component of the median nerve somatosensory evoked potential to be identified. In this way, the multi-stage support vector classifier is designed in a targeted manner, which improves the sensitivity and specificity of classifying and identifying the time-frequency components of somatosensory evoked potentials, and effectively overcomes the problem of small sample size.

Description

technical field [0001] The invention belongs to the technical field of biomedical signal processing, and in particular relates to a recognition method and device of a somatosensory evoked potential component based on a support vector machine. Background technique [0002] Somatosensory evoked potential (SEP) is the electrophysiological response recorded when the nerve is stimulated by the outside world, which can reflect the integrity of the spinal cord and central nervous system. The SEP signal directly reflects the nervous system and contains a large amount of direct key information. Compared with the traditional For radiological examination and neurological examination, the signal recognition technology based on SEP signal components has higher application value. [0003] In the prior art, the recognition and classification of the somatosensory evoked potential components using the artificial graphic labeling method has the disadvantage of taking a long time, and at the s...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, the recognition and classification of the somatosensory evoked potential components using the artificial graphic labeling method has the disadvantage of taking a long time, and at the same time, due to inevitable subjective factors and external interference, it is easy to cause misjudgment

Images