Sleep Apnea Classification Using KNN

Sleep is a state of reduced mental and physical activity, in which consciousness is altered. And all the sensory activity is stopped to certain extent. Maintenance of good sleep is very essential for each individual. Nowadays, diseases related to sleep are very common among the population due to individual living styles and stress related disorder. One such sleep disorder among them is obstructive sleep apnea (OSA), in which an individual experiences difficulty in breathing. Delayed or improper detection of this disorder can cause serious health issues. This work explains the classification of sleep apnea using K-nearest neighbor algorithm using various features. The proposed work is to extract the statistical features such as mean, median, standard deviation and mean square error of ECG signal to detect the abnormal pauses in breathing during sleep apnea period. Classification of sleep stages can be performed using K-nearest neighbor classifier. The KNN classifier was found to work with an accuracy of 95%.