WLnet: Towards an Approach for Robust Workload Estimation Based on Shallow Neural Networks

Zhe Sun,Binghua Li,Cesar F Caiafa,Andrzej Cichocki,Shan Wang,Hao Jia,Feng Duan,Jordi Sole-Casals,Yu Liu

doi:10.1109/access.2020.3044732

Abstract

Electroencephalography (EEG) is a non-invasive technology used for the human brain-computer interface. One of its important applications is the evaluation of the mental state of an individual, such as workload estimation. In previous works, common spatial pattern feature extraction methods have been proposed for the EEG-based workload detection. Recently, several novel methods were introduced to detect EEG pattern workloads. However, it is still unknown which one of these methods is the one that offers the best performance for the workload EEG pattern feature detection. In this article, four methods were used to extract workload EEG features: (a) common spatial pattern feature extraction; (b) temporally constrained sparse group spatial pattern feature extraction; (c) EEGnet; and (d) the new proposed shallow convolutional neural network for workload estimation (WLnet). The classification accuracy of these four methods was compared. Experimental results demonstrate that the proposed WLnet achieved the best detection accuracy in both stress and non-stress conditions. We believe that the proposed methods may be relevant to real-life applications of mental workload estimation.

Highlights

Mental workload estimation-based brain-computer interface has been widely used in many areas, such as in improving the performance of brain-computer interface (BCI), in educational applications or in adapting the difficulty of a task [1]–[5]
Electroencephalography (EEG) signals and near-infrared sensors (NIRS) are usually the technologies used for noninvasive BCI [4], [11]
Some literature has shown that NIRS signals could be used to estimate the mental workload status ( [15]–[17]), in this article we focus on EEG

Summary

Introduction

Mental workload estimation-based brain-computer interface has been widely used in many areas, such as in improving the performance of brain-computer interface (BCI), in educational applications or in adapting the difficulty of a task [1]–[5]. Workload detection has been mainly based on the subjects’ behaviour, eye movement, heart rate and brain activity [6]–[10]. Electroencephalography (EEG) signals and near-infrared sensors (NIRS) are usually the technologies used for noninvasive BCI [4], [11]. EEG equipment consists of metal electrodes which are placed directly on the scalp to record electrical signals [12]. The electrodes record the activity of the surrounding neurons [13]. The other BCI system mentioned previously, the NIRS recording system, is generally used to measure hemodynamic signals from target regions of the brain [14]

Methods

Discussion

Conclusion