Classification of acute stress using linear and non-linear heart rate variability analysis derived from sternal ECG

2014 ◽  
Author(s):  
George Tanev ◽  
Dorthe B. Saadi ◽  
Karsten Hoppe ◽  
Helge B. D. Sorensen
Keyword(s):  
Heart Rate ◽  
Heart Rate Variability ◽  
Acute Stress ◽  
Heart Rate Variability Analysis ◽  
Variability Analysis ◽  
Non Linear

Potential of feature selection methods in heart rate variability analysis for the classification of different cardiovascular diseases

2002 ◽  
Vol 21 (15) ◽  
pp. 2225-2242 ◽  
Author(s):  
Agnes Schumann ◽  
Niels Wessel ◽  
Alexander Schirdewan ◽  
Karl Josef Osterziel ◽  
Andreas Voss
Keyword(s):  
Heart Rate ◽  
Heart Rate Variability ◽  
Feature Selection ◽  
Cardiovascular Diseases ◽  
Heart Rate Variability Analysis ◽  
Selection Methods ◽  
Variability Analysis

The Behavior of Measures of Non-Linear Dynamics in 24 Hours Heart Rate Variability Analysis

1996 ◽  
Vol 91 (s1) ◽  
pp. 122-123
Author(s):  
Niels Wessel ◽  
Andreas Voss ◽  
Juergen Kurths ◽  
Anette Witt ◽  
J Karl Osterziel
Keyword(s):  
Heart Rate ◽  
Heart Rate Variability ◽  
Heart Rate Variability Analysis ◽  
Linear Dynamics ◽  
Variability Analysis ◽  
Non Linear

scholarly journals Reliability of low electrocardiogram sampling frequencies for short-term heart rate variability analysis to estimate transient psychophysiological stress induced by car driving

2020 ◽  
Vol 41 (Supplement_2) ◽  
Author(s):  
D Brisinda ◽  
R Fenici
Keyword(s):  
Heart Rate ◽  
Heart Rate Variability ◽  
Acute Stress ◽  
Intraclass Correlation ◽  
Heart Rate Variability Analysis ◽  
Wireless Devices ◽  
Time Intervals ◽  
Variability Analysis ◽  
Psychophysiological Stress ◽  
The Time Domain

Abstract Background Heart rate variability analysis (HRVa) is increasingly used to evaluate stress-induced adaptation of the autonomic nervous system (ANS), basing on electrocardiogram (ECG) recorded with novel wearable wireless devices (WWD) for physiological monitoring. However, in order to transfer ECG data to mobile phones efficiently, WWD use low sampling frequency (SF), usually ∼100 Hz, thus much lower than the 0.5–1 kHz, previously suggested to optimize R–R intervals detection, precise enough for HRVa in the time-domain (TD), frequency-domain (FD) and with nonlinear (NL) methods. Moreover, the latter are sensitive to SF as a function of data length. Aim Aim of this study was to quantify the relative error (RE%) of HRVa, when calculated from ECG downsampled at 100 Hz, compared with that digitized at 1 kHz, considering different data lengths. Methods ECG of four healthy professional pilots was continuously recorded with a WWD (Vi-grade, Udine), at 1 kHz SF, while driving a professional simulator, inducing different degree of psychophysiological stress by changing suddenly the vehicle's behavior without advising the driver. TD, FD and NL HRV parameters were calculated (Kubios 3.0.2, Finland), from time intervals of 300, 120, 60 and 30 seconds, of ECG downsampled at 100 Hz SF from original recordings acquired at 1 kHz SF (assumed as goldstandard). The RE% and the intraclass correlation coefficient (ICC) were calculated. Results A good correlation (ICC ≥0.79–0.84) was found for the majority of HRV parameters, in both driving conditions, for all selected intervals lengths. The average RE% ranged between zero and 3%, increasing if the length of the interval selected for HRVa was lower than 60 second (Table 1). However higher RE was occasionally found. In Figure 1, an example of HRVa calculated from 2 minutes tachogram's segments, at 100 Hz (left) and at 1 kHz SF (right) is shown. Conclusions Compared to 1 kHz SF, downsampling ECG at 100 Hz doesn't affect significantly HRVa for data lengths between 5 and 1 minutes. However, for shorter time intervals the RE increases. This must be taken into account if HRVa is used to track transient short-lasting changes of ASN modulation induced by acute stress. Figure 1 Funding Acknowledgement Type of funding source: None

Sign in / Sign up

Export Citation Format

Share Document