A Wearable In-Ear EEG Device for Emotion Monitoring

Abstract

For future healthcare applications, which are increasingly moving towards out-of-hospital or home-based caring models, the ability to remotely and continuously monitor patients’ conditions effectively are imperative. Among others, emotional state is one of the conditions that could be of interest to doctors or caregivers. This paper discusses a preliminary study to develop a wearable device that is a low cost, single channel, dry contact, in-ear EEG suitable for non-intrusive monitoring. All aspects of the designs, engineering, and experimenting by applying machine learning for emotion classification, are covered. Based on the valence and arousal emotion model, the device is able to classify basic emotion with 71.07% accuracy (valence), 72.89% accuracy (arousal), and 53.72% (all four emotions). The results are comparable to those measured from the more conventional EEG headsets at T7 and T8 scalp positions. These results, together with its earphone-like wearability, suggest its potential usage especially for future healthcare applications, such as home-based or tele-monitoring systems as intended.

In-Ear EEG Based Attention State Classification Using Echo State Network

“…Since the first research on the “in-the-ear recording concept” was published in 2012 [ 9 ], the BCI application of in-ear EEG signals has been investigated using the external stimuli such as visual or auditory cues [ 11 , 17 , 21 , 22 , 23 ] or independently of external stimuli [ 24 , 25 ]. Compared with the performance of the previous studies on the BCI application of in-ear EEG signals to mental state monitoring, our performance using the ESN technique is higher than theirs: Previous studies successfully have detected drowsiness [ 24 , 25 ], mental workload during visuomotor tracking task [ 26 ], and emotional states [ 27 ] but have required long time window (more than 10 s) to achieve high classification accuracy ( Table 5 ). In this study, we suggest that the attention monitoring system using in-ear EEG and the ESN is much faster to classify mental states than previous studies, within every 0.5 s with high accuracy of 81.16% when using one run as the test set and remaining runs as the training set within each subject.…”

“…Another study reported that mental workload and motor action during a visuomotor tracking task were detected using a two-channel in-ear EEG system with 68.55% accuracy in 5 s windows and 78.51% accuracy when a moving average filter was applied over five such windows [ 26 ]. One study has reported that in-ear EEG signals could be distinguished when subjects viewed emotional pictures for 30 s [ 27 ]. In binary classification tasks, positive valence and negative valence could be discriminated with 71.07% accuracy and high and low arousal could be discriminated with 72.89% accuracy.…”

Emotion Recognition Based on Skin Potential Signals with a Portable Wireless Device

“…With the progress of modern electronics, wearable/portable devices have gradually been developed to collect physiological signals, with the advantages of wearability/portability, wireless capability, and continuous monitoring without causing difficulties in users’ daily lives [ 25 ]. Athavipach et al [ 26 ] discussed a preliminary study to develop a wearable device that is a low-cost, single-channel, dry contact, in-ear EEG suitable for non-intrusive monitoring. The device is able to classify four emotions (happiness, calmness, sadness, and fear) with an accuracy of 53.72%.…”

Physiological Sensors Based Emotion Recognition While Experiencing Tactile Enhanced Multimedia

“…An ensemble classification approach was used to classify emotional states using ECG signals [ 70 ]. Emotion monitoring was proposed for healthcare using a low cost wearable EEG headset [ 71 ]. Moreover, effect of culture on emotion recognition was investigated using EEG signals by presenting video clips in two different languages [ 72 ].…”