Automatic and Accurate Non-contact Obstructive Sleep Apnea Detection using Wavelet Information Entropy Spectrum

Abstract

An accurate non-contact novel detection method for obstructive sleep apnea (OSA) has been developed and will be presented here. Typically, non-contact measurement is preferred for babies, brutally burnt patients, and patients with respiratory contagious diseases, etc. The developed method is based on wavelet information entropy concepts that would clearly classify apnea by the strong abnormality, complex structure and disorder of the patient’s respiratory signal. The accuracy of this method has been experimentally validated, and demonstrated over 93% accuracy using a bio-radar.