Experiments of the sonification of the sleep electroencephalogram

It is becoming possible to perform sleep recordings at home with equipment targeted for the regular consumers. This alleviates the pressures to increase capacity in sleep clinics. The interpretation of the sleep recordings is not very easy for the laymen and alternative assisting methods should be sought for this. Sonification is a method by which a phenomenon is converted to a sound for human listeners. This paper describes experiments made for the sonification of the electric activity of the brain, the electroencephalography (EEG) for the purpose of recognizing the presence and absence of the necessary refreshing components of sleep, deep sleep and rapid eye movement (REM) sleep. The methods are based on the calculation of features of the EEG signal which are characteristic to the deep and REM sleep as well as wakefulness. The features are converted to amplitude modulation functions of artificial and musical instrument sounds by using mathematical transforms such as Principal Component Analysis and Linear Discriminant Analysis. The results indicate that modulated sinusoidal signals are not appropriate for the sonification of sleep EEG but that modulating the sound of musical instruments could be a viable option for making the recognition of good and bad sleep possible