TY - JOUR
T1 - Detrended Fluctuation Analysis Based on the Affective ECG
AU - Jing Cheng & Guangyuan Liu
JO - Journal of Fiber Bioengineering and Informatics
VL - 1
SP - 91
EP - 102
PY - 2014
DA - 2014/07
SN - 7
DO - http://doi.org/10.3993/jfbi03201408
UR - https://global-sci.org/intro/article_detail/jfbi/4769.html
KW - Affective ECG
KW - Press File of Affective Re-evaluation
KW - Detrended Fluctuation Analysis
KW - Scaling Exponent
AB - Electrocardiography (ECG) is one of the most important physiological signals, which has been proven to
contain reliable affective information. Four kinds of objective affects including happiness, sadness, angry
and fear, are induced by affective fragments from movies, and ECG signals are recorded by Biopac MP
150 synchronously. In an independent experiment the affective videos are played twice, and in the second
presentation the press file of affective re-evaluation is obtained, which registers the subjective experience
of participants and help us intercept the reliable affective ECG. The detrended fluctuation analysis is
used to quantitate the temporal correlations by the scaling exponent in affective ECG. And the result
showed that ECG of happiness, sadness, angry and fear had long-range correlations. Then the scaling
exponent is used by the binary classifier of Fisher as an affective feature, and the result showed that
the correct recognition rate of happiness, sadness, angry and fear are 89.74%, 90.1%, 70.43%, 84.44%
respectively. The whole experiment displays that the nonlinear features have a fine distinction in different
emotions.