Eclampsia detection method based on newborn egg signals

Abstract

The invention discloses an eclampsia detection method based on newborn egg signals. The eclampsia detection method comprises a training part and a test part, wherein the training part comprises: performing 0.5-40Hz filtering on data, performing 32Hz down-sampling, performing 0.5-12.5 Hz filtering, segmenting all the data into 24s sections, extracting characteristic values, performing standardizingand performing classification marking on the characteristic values; and performing training by using an SVM to obtain a model. The test part comprises: performing 0.5-40Hz filtering on the data, thenperforming 32Hz down-sampling, then performing 0.5-12.5 Hz filtering, extracting characteristic values from every 24s of data, excluding non-convulsion data according to a threshold condition, standardizing the data which cannot be excluded, performing predicting by using the model, and judging whether the data is convulsion data or not. According to the method, the newborn eeg signals can be subjected to eclampsia detection, medical staff can be assisted in diagnosing neonatal convulsion, and the diagnosis efficiency is improved.

Description

technical field [0001] The invention relates to a method for detecting convulsions based on neonatal electroencephalogram signals, which belongs to image processing technology. Background technique [0002] Convulsion is a temporary disorder of cerebral cortex function that causes abnormal discharge of brain cells. It is manifested as sudden involuntary muscle rigidity and clonic convulsions in the whole body or local skeletal muscles, and causes joint movements. It is mostly systemic and symmetrical, often accompanied by consciousness. obstacle. Convulsions may be due solely to abnormal electrical discharges in the central nervous system, but such abnormal discharges may be caused by many primary intracranial lesions (meningitis, cerebrovascular accident, encephalitis, intracranial hemorrhage, tumors), or secondary to systemic or Metabolic (eg, ischemia, hypoxia, hypoglycemia, hypocalcemia, hyponatremia) disorders. Neonatal convulsions are prone to complications such as a...

AI Technical Summary

Problems solved by technology

[0003] Therefore, the detection of neonatal convulsions is of great significance. At present, there are not many detection equipment for neonatal convulsions. The neonatal EEG is monitored to detect neonatal convulsions. There is no precedent for convulsions

Images