Data extraction method for epilepsy abnormal group activities in interval intracranial electroencephalogram signals

Abstract

The invention discloses a data extraction method for epilepsy abnormal group activities in interval intracranial electroencephalogram signals. According to the method, distribution of epileptiform discharge events along channels and distribution of epileptiform discharge moments in short-time windows are obtained through an automatic detection algorithm of interval epileptiform discharge time, suspicious channels and abnormal time windows are selected according to the distribution, and preliminary screening signals of the epileptiform discharge events are obtained; the primary screening signals can improve the efficiency and quality of analyzing and interpreting the intracranial electroencephalogram signals by electrophysiology personnel; and group activity modes of the epileptiform discharge events are automatically extracted through a non-negative matrix factorization algorithm, the preliminary screening signals of the epileptiform discharge events are classified into different groupactivity modes, and the electrophysiology personnel are assisted in observing group discharge activity modes of multiple channels, so that the data size is reduced, and the utilization rate of the interval intracranial electroencephalogram signals is increased.

Description

Technical field [0001] The invention relates to brain electrical data processing technology, in particular to a data extraction method for epilepsy abnormal group activities in intracranial brain electrical signals during interictal period. Background technique [0002] According to the World Health Organization survey, there are about 50 million people with epilepsy in the world, making epilepsy a common neurological disease. Stereoelectrodeencephalography (SEEG) technology is an intracranial electroencephalogram recorded by deep electrodes, emphasizing the concept of epileptic networks. It was produced in the late 1950s and 1960s by Professors Talairach and Bancaud from the Anne Frank Hospital in Paris, France Since its establishment, SEEG has developed rapidly in the past ten years, and my country has gradually developed SEEG technology since 2012. Clinically, based on the comprehensive analysis of a series of non-invasive examinations such as seizure symptomology, scalp EEG, ...

AI Technical Summary

Problems solved by technology

[0007] 2) Due to the large number of channels of intracranial electrodes, often as high as 100 to 200, it is difficult to observe the synchronization of epileptiform discharge waveforms among multiple channels, and it also leads to the diagnosis that electrophysiologists can obtain from interval data limited information

[0008] The existing invention patent "Intracranial EEG Signal Recognition Device Based on Frequency Band Information and Support Vector Machine" (CN107569228B) provides a method for identifying epileptic intracranial EEG feature waves, but this method is only for spikes, and there is no Provides an automatic identification scheme for waveforms of other classes of epileptiform discharges

[0009] In addition, in the invention patent "method for identifying pathological brain activity based on scalp electroencephalogram" (CN111093471A), it involves a method for automatically detecting multi-channel pathological brain activity patterns with a computer, but it is only used for scalp brain activity. Analysis of electrogram

Images