Method and apparatus for monitoring sleep apnea severity

Disclosed embodiments include a method and an apparatus for monitoring sleep apnea severity that comprise: (a) analyzing an oxygen saturation signal to extract a plurality of time-domain and frequency-domain metrics; (b) calculating an oxygen saturation-based Apnea Hyponea Index (AHI) by employing a predetermined functional mapping between said metrics and polysomnography (PSG)-based AHI; and (c) displaying an oxygen saturation-based AHI to enable a specialist to monitor sleep apnea severity without requiring PSG. According to a particular embodiment, the oxygen saturation signal is nocturnal oxygen saturation and the functional mapping is a multilinear regression model (MLR) or a multilayer perceptron network (MLP).