FIGURE 1 - available via license: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Content may be subject to copyright.
(a), (b) Back and the front side of the ear-hook with behind-the-ear EEG sensors; (c) The proposed device; (d) EEG measurement locations behind the ear.
Source publication
In this study, we propose an end-to-end emotion recognition system using an ear electroencephalogram (EEG)-based on-chip device that is enabled using the machine-learning model. The system has an integrated device that gathers EEG signals from electrodes positioned behind the ear; it is more practical than the conventional scalp-EEG method. The rel...
Contexts in source publication
Context 1
... in the mastoid region positioned posterior to the right ear. These locations were selected due to the ease of signal acquisition compared to on-scalp locations, as well as the reduced susceptibility to interference from extraneous sources such as ocular movements or muscular artifacts. The precise locations of the measurement points are shown in Fig. ...
Context 2
... wearable device (shown in Fig. 1 (c)) combines a high-performance microprocessor for edge machine learning with an accurate analog front-end (AFE) for EEG data acquisition. The device is appropriate for the wearable scenario since it is battery-powered and has an archival size of 6.5 cm 3.5 cm 0.5 cm. Two printed circuit board (PCB) plates were used in the proposed ...
Context 3
... [33] on the relative PSDs of three measuring locations; t-SNE is a method in dimensionality reduction and visualization of highdimensional datasets. Dimensionality reduction typically causes a reduction in the accuracy of the models in classification; therefore, t-SNE is used for data visualization in this study with the SVM classifier (see Fig. 10 (a)). The data point distribution between the two emotional states is considerably pronounced, which proves the feasibility of classifying the collected data after the ...
Context 4
... The application was divided into three major functional groups. The first functional group is concerned with displaying the user's emotional state. The proportion of the relative PSD per frequency band is shown in the second functional group. The third functional group employs audio to notify people immediately when a negative state is detected. Fig. 11 depicts the entire deployment procedure of our real-time EEG-based embedded system for emotion ...
Citations
... The integration of machine learning algorithms with these wearable devices enhances their capabilities by enabling the analysis of complex bio-electrical data to detect anomalies, predict health conditions, and provide personalized recommendations. For instance, machine learning models can be trained to classify ECG signals for blood pressure estimation [6], EEG signals for emotion recognition [16], and EMG signals for gesture recognition [17]. This combination of personalized wearable devices and machine learning holds great promise in advancing preventive healthcare, early disease detection, and personalized treatment strategies, ultimately leading to improved patient outcomes and quality of life. ...
... The real-time emotion recognition system developed by Mai et al. using an ear-EEGbased on-chip device introduces a compact, battery-powered solution for emotion classification. Leveraging machine learning models such as SVM, MLP, and one-dimensional convolutional neural networks (1D-CNNs), this system utilizes Bluetooth low-energy wireless technology for data transmission, showcasing the potential for bio-electrical wearables in mental health applications [16]. Figure 3 represents instances of the examined studies highlighting the utilization of bio-electrical wearable devices employing machine learning techniques. ...
... 2024, 14, x FOR PEER REVIEW 6 of Figure 3 represents instances of the examined studies highlighting the utilization bio-electrical wearable devices employing machine learning techniques. Figure 3. Schematics depicting the sensor setups for: (a) a system to prevent heat stroke in hot en ronments [22], (b) a real-time emotion recognition system [16], and (c) an intelligent wearable sy tem for wound monitoring [28]. (c) is adapted by permission from [28]. ...
This review investigates the convergence of artificial intelligence (AI) and personalized health monitoring through wearable devices, classifying them into three distinct categories: bio-electrical, bio-impedance and electro-chemical, and electro-mechanical. Wearable devices have emerged as promising tools for personalized health monitoring, utilizing machine learning to distill meaningful insights from the expansive datasets they capture. Within the bio-electrical category, these devices employ biosignal data, such as electrocardiograms (ECGs), electromyograms (EMGs), electroencephalograms (EEGs), etc., to monitor and assess health. The bio-impedance and electro-chemical category focuses on devices measuring physiological signals, including glucose levels and electrolytes, offering a holistic understanding of the wearer’s physiological state. Lastly, the electro-mechanical category encompasses devices designed to capture motion and physical activity data, providing valuable insights into an individual’s physical activity and behavior. This review critically evaluates the integration of machine learning algorithms within these wearable devices, illuminating their potential to revolutionize healthcare. Emphasizing early detection, timely intervention, and the provision of personalized lifestyle recommendations, the paper outlines how the amalgamation of advanced machine learning techniques with wearable devices can pave the way for more effective and individualized healthcare solutions. The exploration of this intersection promises a paradigm shift, heralding a new era in healthcare innovation and personalized well-being.
