Efficient wavelet-based artifact removal for electrodermal activity in real-world applications

Online monitoring of electrodermal activity (EDA) may serve as an economical and explicit source of information about actual emotional state and engagement level of users during their interaction with information and communications technologies (ICT) applications in real-world situations. In such contexts, however, EDA signal is affected by motion artifacts that introduce noise in the signal and can make it unusable. As the scope of movement minimization during EDA data acquisition is limited, this scenario demands online methods for detection and correction of artifacts with low computational cost. We propose an efficient wavelet-based method for artifacts attenuation while minimizing distortions, using a stationary wavelet transform (SWT) modeling the wavelet coefficients as a Laplace distribution. The proposed method was tested on EDA recordings from publicly available driver dataset collected during real-world driving, and containing a high number of motion artifacts, and the results were compared to those of three state-of-the-art methods for EDA signal filtering. In addition, the proposed method was tested for the online filtering of EDA signals collected while 12 volunteers conducted tasks designed to elicit various stress states. The results evidenced that the prediction of arousal states can be significantly improved after motion artifacts removal, and that the proposed method outperforms existing approaches and it has a lower computational cost. Taken together, these results evidence the effectiveness of the proposed method for online EDA filtering in real world scenarios.