Electroencephalogram pattern recognition using fuzzy logic

A fuzzy logic approach to the classification of human sleep using electroencephalogram (EEG) data is presented. In this approach, frequency and amplitude information from an epoch of the EEG signal are extracted into a vector that is then compared to previously taught vectors representing the canonical features of six stages: wakefulness, rapid eye movement (REM) sleep, and four nonREM sleep stages. For each stage, membership functions are calculated in each epoch. The stage with the maximum degree of membership is scored and classified. The system is implemented in software using the C programming language. Analysis of about 1101 epochs of the EEG data yielded an overall agreement of 77% between the program and a human scorer