GC-106 - Emotion Recognition Using Wireless Signals

Stress can increase the heart rate, causing dangerous conditions that cause significant harm and even death. Therefore, managing stress well to control the heart rate is vital. Monitoring stress levels using biomedical devices is essential to prevent further damage to the individual. Regular monitoring facilitates the acquisition of individual stress levels through heart rate information. In this research, a MAX30100 sensor can produce Beats Per Minute (BPM) data, which show the stress experienced by a person during four daily activities (sitting, walking, running, and sleeping) and when watching a horror movie. This research underscores the potential of cost-effective wireless solutions to increase portability and facilitate real-time monitoring and remote surveillance. The researchers introduced an innovative methodology by reusing everyday consumer devices, initially intended for routine activities, to measure stress levels in a unique context-precisely in a controlled experiment involving a movie-watching scenario. The ultimate achievement of this research was creating a stationary heart rate monitoring device that is easy to access, easy to use, space-saving, and adept at effectively tracking stress indicators. BPM readings obtained from the MAX30100 sensor have scores similar to those obtained from the heart rate app. The researchers proposed the inclusion of other sensors, such as a photoplethysmograph, an electrocardiogram, and galvanic skin response, to provide a more holistic assessment of stress. Leveraging cloud technology for data storage enables long-term monitoring and in-depth analysis of stress patterns over time.

Emotions are fundamental to humans. They affect perception and everyday activities such as communication, learning and decision making. Various emotion recognition devices have been developed incorporating facial expressions, body postures and speech recognitions as a means of recognition. The accuracy of most of the existing devices is dependent on the expressiveness of the user and can be fairly manipulated. We proposed a physiological signal based solution to provide reliable emotion classification without possible manipulation and user expressiveness. Electrocardiogram (ECG) and Galvanic Skin Response (GSR) signals are extracted using shimmer sensors and are used for recognition of seven basic human emotions (happy, fear, sad, anger, neutral, disgust and surprise). Classification of emotions is performed using Convolutional Neural Network. Using AlexNet architecture and ECG signals, emotion classification accuracy of 91.5% for AMIGOS dataset and 64.2% for a real-time dataset is achieved. Similarly, the accuracy of 92.7% for AMIGOS dataset and 68% for a real-time dataset is achieved using GSR signals. Through combining both ECG and GSR signals the accuracy of both, AMIGOS and real-time datasets is improved to 93% and 68.5% respectively.

IEEE Transactions on Affective Computing

Smart wearables, equipped with sensors monitoring physiological parameters, are becoming an integral part of our life. In this work, we investigate the possibility of utilizing such wearables to recognize emotions in the wild. In most reviewed papers, the authors apply a similar procedure consisting of participant recruitment, stimuli preparation and annotation, signal collection and processing, self-assessment, and machine learning model learning and validation. Besides, we identified seven emotion recognition scenarios and analyzed the transition from psychological models to machine learning tasks. Even though the majority of the research was performed in the laboratory environment, we conclude that studies in the field are feasible. They require especially: (1) new self-assessment and triggering procedures adjusted to a real-life scenario, (2) more attention to the machine learning process, including suitable deep learning architectures, revision of the data imbalance problem, and subject-specific data processing, (3) adequate validation procedures, (4) consideration of the model generalizability vs. personalizability, (5) comfortable devices able to provide reliable measurements in motion. Additionally, more large-scale studies are necessary to increase result credibility. We also postulate actions toward replicability and comparability of the research.

9th International Conference on Affective Computing and Intelligent Interaction Workshops and Demos (ACIIW), 2021

Several obstacles have to be overcome in order to recognize emotions and affect in daily life. One of them is collecting a large amount of emotionally annotated data necessary to create data-greedy machine learning-based predictive models. Hence, we propose the Emognition system supporting the collection of rich emotional samples in everyday-life scenarios. The system utilizes smart-wearables to record physiological signals unobtrusively and smartphones to gather self-assessments. We have performed a two-week pilot study with 15 participants and devices available on the market to validate the system. The outcomes of the study, alongside the discussion and lessons learned, are provided.

Various instrumental methods have been developed for observing human emotions in psychology, the neurosciences, and machine learning studies. These methods are mostly directed to detecting individual, personal emotions. The focus of this study is developing instrumental methods for remote detection of human emotions, both individual and collective. A new antenna device has been added to the Electrophotonic Imaging/Gas Discharge Visualization camera system. Examples of data are explored from a water blessing ritual, healing workshop training, musical performance, geophysical measurements during a solar eclipse, and a shamanic ceremony. These preliminary indications demonstrate a potential for mapping the effects of human emotions. We propose an international collaborative project for a Geoactive Zones Database: developing maps of energy parameters for sacral subjects and a program for study of their influence on the human psycho-physiological condition; in relation with both the envi...

EURASIP Journal on Advances in Signal Processing, 2004

We discuss the strong relationship between affect and cognition and the importance of emotions in multimodal human computer interaction (HCI) and user modeling. We introduce the overall paradigm for our multimodal system that aims at recognizing its users' emotions and at responding to them accordingly depending upon the current context or application. We then describe the design of the emotion elicitation experiment we conducted by collecting, via wearable computers, physiological signals from the autonomic nervous system (galvanic skin response, heart rate, temperature) and mapping them to certain emotions (sadness, anger, fear, surprise, frustration, and amusement). We show the results of three different supervised learning algorithms that categorize these collected signals in terms of emotions, and generalize their learning to recognize emotions from new collections of signals. We finally discuss possible broader impact and potential applications of emotion recognition for multimodal intelligent systems.

Biosensors

Automatically recognizing negative emotions, such as anger or stress, and also positive ones, such as euphoria, can contribute to improving well-being. In real-life, emotion recognition is a difficult task since many of the technologies used for this purpose in both laboratory and clinic environments, such as electroencephalography (EEG) and electrocardiography (ECG), cannot realistically be used. Photoplethysmography (PPG) is a non-invasive technology that can be easily integrated into wearable sensors. This paper focuses on the comparison between PPG and ECG concerning their efficacy in detecting the psychophysical and affective states of the subjects. It has been confirmed that the levels of accuracy in the recognition of affective variables obtained by PPG technology are comparable to those achievable with the more traditional ECG technology. Moreover, the affective psychological condition of the participants (anxiety and mood levels) may influence the psychophysiological respon...

ICTACT Journal on Communication Technology

Emotion recognition is one of the biggest challenges in human-human and human-computer interaction. There are various approaches to recognize emotions like facial expression, audio signals, body poses, and gestures etc. Physiological signals play vital role in emotion recognition as they are not controllable and are of immediate response type. In this paper, we discuss the research done on emotion recognition using skin conductance, skin temperature, electrocardiogram (ECG), electromyography (EMG), and electroencephalogram (EEG) signals. Altogether, the same methodology has been adopted for emotion recognition techniques based upon various physiological signals. After survey, it is understood that none of these methods are fully efficient standalone but the efficiency can be improved by using combination of physiological signals. The study of this paper provides an insight on the current state of research and challenges faced during emotion recognition using physiological signals, so that research can be advanced for better recognition.

2017 Seventh International Conference on Affective Computing and Intelligent Interaction (ACII), 2017

Emotion recognition from biometrics is relevant to a wide range of application domains, including healthcare. Existing approaches usually adopt multi-electrodes sensors that could be expensive or uncomfortable to be used in real-life situations. In this study, we investigate whether we can reliably recognize high vs. low emotional valence and arousal by relying on noninvasive low cost EEG, EMG, and GSR sensors. We report the results of an empirical study involving 19 subjects. We achieve state-of-theart classification performance for both valence and arousal even in a cross-subject classification setting, which eliminates the need for individual training and tuning of classification models.

Diagnostics

Emotion recognition becomes an important aspect in the development of human-machine interaction (HMI) systems. Positive emotions impact our lives positively, whereas negative emotions may cause a reduction in productivity. Emotionally intelligent systems such as chatbots and artificially intelligent assistant modules help make our daily life routines effortless. Moreover, a system which is capable of assessing the human emotional state would be very helpful to assess the mental state of a person. Hence, preventive care could be offered before it becomes a mental illness or slides into a state of depression. Researchers have always been curious to find out if a machine could assess human emotions precisely. In this work, a unimodal emotion classifier system in which one of the physiological signals, an electrocardiogram (ECG) signal, has been used is proposed to classify human emotions. The ECG signal was acquired using a capacitive sensor-based non-contact ECG belt system. The machi...