For automatic detection and monitoring of obstructive sleep apnea

Abstract

Nonlinear/chaotic algorithms are used for automatic detection and clinical monitoring of obstructive sleep apnea (OSA). We cite an example taken from a group of adults where similar results are obtained. The algorithms are applied to clinical time series of airflow (thermistry), chest effort (impedance) and electrocardiogram (ECG) traces obtained from sleep apnea records (Edentrace Systems). These algorithms may be applied to a variety of other data sets (i.e. oxygen saturation, heart rate). The algorithms can pinpoint the onset of a disabling disorder (i.e apnea) and mark the duration of the event. They are robust when applied to multiple data sets in which obstructive apneas are known to occur.