| Abstract [eng] |
Stress is a fairly common health problem associated with increased morbidity and mortality from cardiovascular disease. Parameters of heart rate variability are most commonly used to evaluate it. However, the evaluation of these parameters requires a long time window, which leads to long time delays in feedback system and other inconveniences for user of system. In this work, it is proposed to use alternative, faster-changing stress level markers evaluated from photoplethysmogram (PPG) signals. In order to manage and reduce stress, biological feedback could be applied as slow deep breathing at the resonant frequency. In this work, seven pulse wave features of demodulated PPG signal were evaluated – systolic and diastolic areas of PPG pulse As and Ad, slope coefficients Sb-c and Sb-d, time intervals Ta-b, Tb-c and Tbd. In addition, the pulse amplitude variability PAV of PPG signal and pulse duration PP were estimated. Respiratory frequency was calculated based on the analysis of PPG signals. A database of signals from 51 subject was registered to investigate the psychophysiological response of the features to the short-term physical stress and deep breathing. Subjects were stressed using a cold pressor test. The signals were registered with the device "Nautilus 2.0" developed by the staff of the Institute of Biomedical Engineering of Kaunas University of Technology. Statistical analysis of PPG signal parameters revealed that the most physical stress and breathing sensitive parameter is the pulse amplitude variability PAV (Cohen’s d values – 1,444; 1,129; 1,030). Therefore, biological feedback in the system could be implemented based on this parameter with the possibility to additionally monitor other signal parameters. This type of stress management system has a potential to help reduce a pain caused by physical stress, which is often felt by people who have experienced severe physical injuries. A rather accurate algorithm for estimation of the respiratory resonant frequency has been implemented (median - 6,229 breaths/min). However, the influence of deep breathing on pulse wave features of PPG signal should be investigated in more details in the future, because not all parameters responded sensitively to the stimulus of deep breathing. It may be possible to increase the duration of deep breathing phase, adjust inhalation, exhalation, and holdup periods, or perform exhalation with the mouth open. |
