RECOGNITION OF RAPID-EYE-MOVEMENT SLEEP FROM SINGLE-CHANNEL EEG DATA BY ARTIFICIAL NEURAL NETWORKS - A STUDY IN DEPRESSIVE PATIENTS WITH AND WITHOUT AMITRIPTYLINE TREATMENT

An automatic procedure for the online recognition of REM sleep appears to be a necessary tool for selective REM sleep deprivation in depress ive patients. To develop such a procedure we applied an artificial neu ral network to preprocessed single-channel EEG activity. EOG and EMG i nformation was purposely not provided as input to the network. A gener alized back-propagation algorithm was used for computer simulation. Th e sleep profile scored manually according to Rehtschaffen and Kales se rved as the desired output during the training period and as standard for the judgement of the network output during working mode. Polysomno graphic recordings from 5 healthy subjects were pooled to train the ne twork, whereas second-night EEG recordings from the same subjects were used as independent working data sets. We further applied the network to the data of 5 depressive patients without medication and 6 depress ive patients treated with amitriptyline. For these groups between 84.9 and 88.6% out of all time periods consisting of 20 s of continuous EE G activity were correctly classified. The indicator function of REM sl eep was well approximated by the network output in the course of the n ight. Especially the REM onset was excellently recognized. The inclusi on of patient data in the training set yielded a different network, wh ich was evaluated and compared.