An Approach for Prediction of Obstructive Sleep Apnea

Abstract

Obstructive Sleep Apnea (OSA) is a common disorder related to breathing, happens when obstruction occurred due to physical collapse of pharyngeal airways. When obstructive events happens over and over in the aviation route during rest due to unwinding of the tongue and aviation route muscles then it results in OSA. . Regular markers of OSA are snore throat at wake ups ,dry mouth, poor night sleep and morning headaches. Determination of OSA is expensive in terms of time and money. So the patients not easily discovered and so they become unconscious for their condition. Obstructive sleep apnea is a typical issue that influences a person’s breathing during sleep. Sleep apnea is mainly of two types, but most popular type is Obstructive Sleep Apnea(OSA). In last few decades an Sleep Apnea especially Obstructive sleep apnea (OSA) is progressively becomes as an important medical problem.

This paper analyzes the utilization of profound learning methods through the profundity guide of human facial outputs instead of plain 2-D shading picture using morphological operations. So in order to simplify diagnosis or detection of OSA, we found that using morphological operations on parameters like facial width, distance between eyes and chin-neck angle, we can make prediction of OSA.