Automatic recognition of sleep spindles using short-term spectral analysis

An approach tö recognize automatically sleep spindles by using short-term spectral analysis is described. The EEG signal is sampled and stored on a magnetic tape in digital form. The data is analyzed on a PDP-8 computer, and the result can be either displayed on an oscilloscope or stored on another magnetic tape for further analysis. The program is written in conversational form and provides the operator with the ability to change any parameter in the program quickly by a simple command on the Teletype keyboard.