Respiratory scoring using photoplethysmography signal and nearest neighbor algorithm

Obstructive sleep apnea is diagnosed by devices using polysomnography method. Some standard parameters are used for measurement. Diagnosis is made by a specialist doctor with respiratory scoring and sleep staging through parameters. The need for a new system arises because of the cost of the polysomnography method, the need for laboratories and specialists to implement it, the lack of measurement reliability and the delay in diagnosis. It is considered that a new system to be constructed using the Photoplethysmograohy (PPG) signal will meet this need. It was aimed to perform respiratory scoring with the help of classifying PPG signal with k nearest neighborhood algorithm. The results were obtained for 5 different values of k and the best result was obtained with k = 6 value with 88.6% accuracy. As a result, PPG signal could be used as an alternative method for respiratory scoring.