Statistical investigation of heart rate variable derived from photoplethysmography signal in sleep staging process

Abstract

Obstructive Sleep Apnea diagnosis is diagnosed by a specialist doctor, examining the records taken with the polysomnography device. Sleep staging is the first and most important step in diagnosis. In the second step, diagnose with respiratory scoring. Sleep staging can be performed with at least three signals and ten channel electrodes. The electrodes are can be connected only by a qualified technician to the patient. Electrode connection patterns disrupt the patient's natural sleep environment. In addition, diagnosis can be made only in the laboratory. There is a need for practical measurement systems to solve these problems. In this study, it is suggested that Heart Rate Variable (HRV) derived from Photopethemography (PPG) signal can be used in sleep staging. In the study, HRV was derived from the PPG signal. 15 features were extracted from the HRV in the time domain. Mann-Whitney U Test was used for statistical analysis and distinctive features have been identified for sleep arousal. Mann-Whitney U Test was used to determine whether the extracted features were a distinguishing feature for the Sleep Awakening status. When the results are examined, it was determined that p < 0.05 for all features discriminant for sleep wakefulness. When the results are examined, HRV derived from the PPG signal can be used to develop a sleep staging system which can be used at home.