Method for real time automatically detecting epileptic character wave

Abstract

The invention discloses a method for detecting epileptic characteristic waves automatically and in real-time by using methods of empirical mode decomposition, Hilbert transformation and a smooth nonlinear energy operator. The real-time and automatic detection of the epileptic characteristic waves needs to divide brain wave data into sections by using a movable window technology, and calculating the brain wave data in each section by methods as follows: firstly, original brain wave signals are divided into a series of inherent mode functions IMF; then, the inherent mode functions IMF with high signal to noise ratio is selected and preprocessed by using an automatic wavelet transform noise-removing method; transient energy of each inherent mode function is calculated by using the Hilbert transformation and total transient energy S is calculated by summing; the smooth nonlinear energy operator is used for detecting spikes of the total transient energy S. The method is applicable to retrieve the epileptic characteristic waves of the brain waves automatically and in real-time, which has the important meanings in diagnoses for patients and reducing heavy work of doctors.

Description

technical field [0001] The invention relates to a method for automatically detecting epileptic seizures in real time, in particular to a method for detecting brain epilepsy spikes. Background technique [0002] EEG is the most important auxiliary examination method for the diagnosis of epilepsy. It can help doctors determine whether the patient's seizures are epilepsy, especially for those diseases with atypical seizures or similar seizures, which can help in differential diagnosis. Many clinical diseases, such as apnea, movement disorder, syncope, arrhythmia, sleep disorder, migraine and various neurological symptoms, sometimes very similar to epileptic seizures, we can often make accurate diagnoses by virtue of EEG examination. Most patients with clinically typical epileptic seizures can find epileptiform characteristic waves in EEG examination. The epileptiform characteristic waves mainly include the following types: spike (or sharp) wave, spike (or sharp) slow complex, ...

AI Technical Summary

Problems solved by technology

Its disadvantages are: due to the complexity of the epilepsy waveform, it is difficult to define a template set suitable for common cases, and it is also difficult to set the threshold in order to find a compromise between the false detection rate and the missed detection rate

Therefore, whether it is an automatic detection method in the time domain based on EEG waveform analysis or an automatic detection method in the frequency domain using a parametric model, there are inevitable difficulties in practical application.

Images