Quantify stress levels using psycho-physiological signals
Quantify stress levels using psycho-physiological signals
mehran nematollahi1
1) MSc in Biomedical Engineering, Islamic Azad University, Ahar Branch, Iran -
IOI Code
:
XBAA-DZEEH
Publication :
International Congress on Engineering science and Sustainable Urban Development - Copenhagen (dtuconf.eu)
Abstract :
In psychopysiology research, we give a psychological sense to the responses of physiology of the body, with regard to factors such as the design quality of the test, the psychological properties of the measurements, and the appropriateness of the analysis and interpretation of the data. In analyzing reactions, none of the two physiology and psychology sciences are superior, but complement each other. Recognizing various mental states, including the state of stress, which has a known destructive effect on human body and soul, is one of the important applications of this science. In this paper, by providing a proper test and creating three levels of stress (low, moderate, and high) in the subject and recording peltomisograms, changes in heart rate and electrical conductivity of the skin, we have sought to obtain a standard for quantifying the level of stress in the individual. For this purpose, various linear processing and processing in the time domain, frequency and time - frequency and nonlinear, including Poanchar, Lyapunov exponant, fractal dimension and entropy, and extraction of various characteristics of the recorded signals have been made. Then, by applying different classification methods, including composition of the neural network and the genetic algorithm, backup vector machines and the method of linear composition function are differentiated into different levels. In this paper, the optimal characteristics of each signal were first determined and differentiated into these three methods. Then, by combining the optimal characteristics of all the signals, the separation was performed again. Finally, it was concluded that using the HRV signal alone, higher resolution results could be achieved. In the following, a comparison was made between the linear and nonlinear characteristics of the HRV signal and it was concluded that the combination of these two types of characteristics improved the results.
Keywords : peltisogramm signals heart rate poanchar criteria lyapunov exponent hrv signal