Fast Stress Detection via ECG
MetadataShow full item record
Nowadays stress has become a regular part of life. Stress is difficult to measure because there has been no definition of stress that everyone accepts. Furthermore, if we do not get a handle on our stress and it becomes long term, it can seriously interfere with our health. Therefore, finding the method for stress detection could be beneficial for taking control of stress. Electrocardiogram (ECG) is the measurement of the electrical activity of the heart and represents an established standard in determining the health condition of the heart. The PQRST1 complex of ECG conveys information about each cardiac-cycle, where the R-peak is placed in the middle of the PQRST complex and represents the maximum value of the PQRST. Since the PQRST depicts the entire cardio-cycle, the R–peak determines half of the cardio-cycle. The distance between two adjacent R-peaks is defined as a heart rate (HR). The variation of the HR in the specific time frame, defined as heart rate variability (HRV), can reflect the state of the autonomic nervous system (ANS). The ANS has two main divisions, the sympathetic nervous system (SNS) and the parasympathetic nervous system (PNS). The SNS occurs in response to stress while the PNS results from the function of internal organs. The activity of ANS can cause an acceleration (SNS) or deceleration (PNS) of the HR. The SNS activity is associated with the low-frequency range while, the PNS activity is associated with the high frequency component of the HRV. Therefore, the power ratio of the low and high-frequency components of the spectrum of HRV can potentially show whether the subject is exposed to stress or not  . In this research, we introduced three new indices, with one of them proposed as a proxy to provide equivalent results in the detection of stress or no-stress states while avoiding complex measurement devices as well as complex calculations. The goal was to find a more time efficient method for fast stress detection which could potentially be used in the applications that run on devices such as a wearable smartwatch in tandem with a smartphone or tablet. The experiment was established to measure the literature proposed index for stress measurement  as well as our introduced indices. In the experiment, we induced stress to the participants by using mental arithmetic as a stressor . Theexperiment contained two kinds of trials. In the first one, the participant was exposed to different amounts of cognitive load induced by doing mental-arithmetic while, in the second one, the participant was placed in a relaxed environment. Each participant in the experiment gave feedback in which period of the experiment he/she felt stress. During the entire experiment, we recorded theparticipant‘s ECG. The ECG was used to calculate HRV which consequently was used for the calculation of the values of the index as proposed from the literature for calculating the level of the stress. The same data was used for the calculation of our introduced indices. The values of our proposed index was compared with the index and the participant‘s feedback. Finally, the data analyses showed that our proposed index is suitable to determine whether a participant is exposed to stress.
Cite this version of the work
Aleksandar Malinovic (2019). Fast Stress Detection via ECG. UWSpace. http://hdl.handle.net/10012/14698