The Research And Design Of Electroencephalogram Auto-Detection And Intelligent Diagnosis System

EEG has been widely used in diagnosis of brain disease and neural system disease.Intelligent detection and diagnosis of EEG have good use in clinical treatment, especially for the epileptic symptom.Though there are some approaches to solving the problem of auto electroencephalography (EEG) analysis, and yet many methods can't deal with abnormal signal efficiently, leading to poor results.For this, the paper improved the effect of auto-detection by a series of method. First,collected data,filtered waves,and extracted features,so the wave parameters can be obtained. Then waves are to be classified by inputting these parameters to neural network, using NNCA (Nearest-Neighbor Clustering Algorithm) to train it, At last we applied the whole process to the experiment of detecting epileptic signal. In finally, expert system makes diagnosis on disease according to disease symptoms.It can help the doctors to make diagnosis.The system can improve the diagnosis accuracy by accumulating samples in clinical treatment. It process the signal in several hierarchical stages and utilizes various digital signal processing methods including wavelet transform, artificial neural network and expert system.The research idea can also be applied to other medical diagnosis systems with bioelectricity.Test on clinical recordings indicates the system has a detection Rate greater than 80%.The hierarchical multi-method strategy improves the system sensitivity and selectivity while greatly reducing the computational load.So it can fulfill the need for long term electroencephalography (EEG) data processing.