Metric multidimensional scaling and aggregation operators for classifying epilepsy from EEG signals

Abstract

As a result of sudden and excessive electrical discharges in a specific group of brain cells called neurons, epilepsy occurs and is usually for a brief period. It can occur in various parts of the brain and the patient can experience different symptoms depending on the occurrence of the excessive discharges. So the electrical impulses generated due to the nerve firing in the brain can be measured easily with the help of Electroencephalogram (EEG) by placing the electrodes on the scalp of the patient. As the recordings are too long, the data to be processed is large and hence Metric Multidimensional Scaling (MDS) is used to reduce the dimensions of the EEG data. The dimensionally reduced values are then fed inside the Aggregation Operator Classifiers to classify the epilepsy from EEG signals. Results show that an average accuracy of 92.36% along with an average time delay of 2.44 seconds is found out.