1st International and 10th National Iranian Conference on Bioinformatics
Home Accepted Papers
A greedy time-frequency analysis of electrodermal activity for cognitive state monitoring
Paper ID : 1216-ICB10
Authors:
Rezvan Mirzaeian1, پیوند قادریان *2
1دانشگاه صنعتی سهند تبریز
2دانشکده مهندسی پزشکی، دانشگاه صنعتی سهند، تبریز، ایران
Abstract:
Cognitive workload (CW) is defined as the required mental effort to perform a task, which is associated with capacity-limited cognitive system for processing information in the working memory [1]. The development of a framework to not only minimize human errors, but also maximize performance, is an important application of CW estimation and management [2]. It can also help to improve the diagnosis and treatment of neurological or cognitive disorders, and to facilitate human-computer interaction [3]. In place of non-invasive nature, sensitivity to cognitive states, robustness against intentional conduct, and the possibility of online investigation, psychophysiological signal analysis have received special attention for CW estimation [4]. Hence, a novel CW estimation method based on matching pursuit (MP) decomposition of electrodermal activity (EDA) and support vector machine classifier has been proposed. The MP, as adaptive and greedy time-frequency decomposition, has the advantages of reducing cross terms, improving time-frequency resolution and enhancing biomedical signal analysis performance [5, 6]. Applying two dictionaries, including RnIdent and daubechies wavelet (db5) at level 5, sparse coefficients have been extracted from the EDA signals. Then, several statistical and nonlinear features (mean, standard deviation, variance, covariance, and Shannon entropy) have been calculated from the MP coefficients. The proposed method evaluated on EDA of 30 healthy students performing an arithmetic task has achieved an average accuracy of 96.80% for two workload levels. Moreover, the experimental results have indicated that the combination of complementary information from the different extracted features has enhanced the estimation performance.
Keywords:
Matching Pursuit, Cognitive workload, Support vector machine, arithmetic task
Status : Paper Accepted (Poster Presentation)