Search

Despite the notable recent developments in the field of remote photoplethysmography (rPPG), extracting a reliable pulse rate variability (PRV) signal still remains a challenge. In this study, eight image-based photoplethysmography (iPPG) extraction methods (GRD, AGRD, PCA, ICA, LE, SPE, CHROM, and POS) were compared in terms of pulse rate (PR) and PRV features. The algorithms were made robust for motion and illumination artifacts by using ad hoc pre- and postprocessing steps. Then, they were systematically tested on the public dataset UBFC-RPPG, containing data from 42 subjects sitting in front of a webcam (30 fps) while playing a time-sensitive mathematical game. The performances of the algorithms were evaluated by statistically comparing iPPG-based and finger-PPG-based PR and PRV features in terms of Spearman’s correlation coefficient, normalized root mean square error (NRMSE), and Bland–Altman analysis. The study revealed POS and CHROM techniques to be the most robust for PR estimation and the assessment of overall autonomic nervous system (ANS) dynamics by using PRV features in time and frequency domains. Furthermore, we demonstrated that a reliable characterization of the vagal tone is made possible by computing the Poincaré map of PRV series derived from the POS and CHROM methods. This study supports the use of iPPG systems as promising tools to obtain clinically useful and specific information about ANS dynamics.

[EN] Virtual reality is a powerful tool in human behaviour research. However, few studies compare its capacity to evoke the same emotional responses as in real scenarios. This study investigates psycho-physiological patterns evoked during the free exploration of an art museum and the museum virtualized through a 3D immersive virtual environment (IVE). An exploratory study involving 60 participants was performed, recording electroencephalographic and electrocardiographic signals using wearable devices. The real vs. virtual psychological comparison was performed using self-assessment emotional response tests, whereas the physiological comparison was performed through Support Vector Machine algorithms, endowed with an effective feature selection procedure for a set of state-of-the-art metrics quantifying cardiovascular and brain linear and nonlinear dynamics. We included an initial calibration phase, using standardized 2D and 360 degrees emotional stimuli, to increase the accuracy of the model. The self-assessments of the physical and virtual museum support the use of IVEs in emotion research. The 2-class (high/low) system accuracy was 71.52% and 77.08% along the arousal and valence dimension, respectively, in the physical museum, and 75.00% and 71.08% in the virtual museum. The previously presented 360 degrees stimuli contributed to increasing the accuracy in the virtual museum. Also, the real vs. virtual classifier accuracy was 95.27%, using only EEG mean phase coherency features, which demonstrates the high involvement of brain synchronization in emotional virtual reality processes. These findings provide an important contribution at a methodological level and to scientific knowledge, which will effectively guide future emotion elicitation and recognition systems using virtual reality.

[EN] Affective Computing has emerged as an important field of study that aims to develop systems that can automatically recognize emotions. Up to the present, elicitation has been carried out with nonimmersive stimuli. This study, on the other hand, aims to develop an emotion recognition system for affective states evoked through Immersive Virtual Environments. Four alternative virtual rooms were designed to elicit four possible arousal-valence combinations, as described in each quadrant of the Circumplex Model of Affects. An experiment involving the recording of the electroencephalography (EEG) and electrocardiography (ECG) of sixty participants was carried out. A set of features was extracted from these signals using various state-of-the-art metrics that quantify brain and cardiovascular linear and nonlinear dynamics, which were input into a Support Vector Machine classifier to predict the subject's arousal and valence perception. The model's accuracy was 75.00% along the arousal dimension and 71.21% along the valence dimension. Our findings validate the use of Immersive Virtual Environments to elicit and automatically recognize different emotional states from neural and cardiac dynamics; this development could have novel applications in fields as diverse as Architecture, Health, Education and Videogames.

[EN] Affective Computing has emerged as an important field of study that aims to develop systems that can automatically recognize emotions. Up to the present, elicitation has been carried out with nonimmersive stimuli. This study, on the other hand, aims to develop an emotion recognition system for affective states evoked through Immersive Virtual Environments. Four alternative virtual rooms were designed to elicit four possible arousal-valence combinations, as described in each quadrant of the Circumplex Model of Affects. An experiment involving the recording of the electroencephalography (EEG) and electrocardiography (ECG) of sixty participants was carried out. A set of features was extracted from these signals using various state-of-the-art metrics that quantify brain and cardiovascular linear and nonlinear dynamics, which were input into a Support Vector Machine classifier to predict the subject's arousal and valence perception. The model's accuracy was 75.00% along the arousal dimension and 71.21% along the valence dimension. Our findings validate the use of Immersive Virtual Environments to elicit and automatically recognize different emotional states from neural and cardiac dynamics; this development could have novel applications in fields as diverse as Architecture, Health, Education and Videogames.

Mental diseases are increasingly common. Among these, bipolar disorders heavily affect patients’ lives given the mood swings ranging from mania to depression. Voice has been shown to be an important cue to be investigated in relation with this kind of disease. In fact, several speech-related features have been used to characterize voice in depressed speakers. The goal is to develop a decision support system facilitating diagnosis and possibly predicting mood changes. Recently, efforts were devoted to studies concerning bipolar patients. A spectral analysis of F0-contours extracted from audio recordings of a text read by bipolar patients and healthy control subject is reported. The algorithm is automatic and the obtained features describe parsimoniously speech rhythm and intonation. Bipolar patients were recorded while experiencing different mood states, whereas the control subjects were recorded at different days. Feature trends are detected in bipolar patients across different mood states, while no significant differences are observed in healthy subjects., SCOPUS: ar.j, info:eu-repo/semantics/published

info:eu-repo/semantics/published

info:eu-repo/semantics/published

Emotion recognition through computational modeling and analysis of physiological signals has been widely investigated in the last decade. Most of the proposed emotion recognition systems require relatively long-time series of multivariate records and do not provide accurate real-time characterizations using short-time series. To overcome these limitations, we propose a novel personalized probabilistic framework able to characterize the emotional state of a subject through the analysis of heartbeat dynamics exclusively. The study includes thirty subjects presented with a set of standardized images gathered from the international affective picture system, alternating levels of arousal and valence. Due to the intrinsic nonlinearity and nonstationarity of the RR interval series, a specific point-process model was devised for instantaneous identification considering autoregressive nonlinearities up to the third-order according to the Wiener-Volterra representation, thus tracking very fast stimulus-response changes. Features from the instantaneous spectrum and bispectrum, as well as the dominant Lyapunov exponent, were extracted and considered as input features to a support vector machine for classification. Results, estimating emotions each 10 seconds, achieve an overall accuracy in recognizing four emotional states based on the circumplex model of affect of 79.29%, with 79.15% on the valence axis, and 83.55% on the arousal axis., Harvard Medical School, Massachusetts General Hospital. Department of Anesthesia, Critical Care & Pain Medicine, Seventh Framework Programme (European Commission) (FP7/2007–2013 under grant agreement n 601165 of the project “WEARHAP”), Seventh Framework Programme (European Commission) (FP7/2007–2013 under grant agreement ICT-247777 of the project “PSYCHE”)

Measures of entropy have been widely used to characterize complexity, particularly in physiological dynamical systems modeled in discrete time. Current approaches associate these measures to finite single values within an observation window, thus not being able to characterize the system evolution at each moment in time. Here, we propose a new definition of approximate and sample entropy based on the inhomogeneous point-process theory. The discrete time series is modeled through probability density functions, which characterize and predict the time until the next event occurs as a function of the past history. Laguerre expansions of the Wiener-Volterra autoregressive terms account for the long-term nonlinear information. As the proposed measures of entropy are instantaneously defined through probability functions, the novel indices are able to provide instantaneous tracking of the system complexity. The new measures are tested on synthetic data, as well as on real data gathered from heartbeat dynamics of healthy subjects and patients with cardiac heart failure and gait recordings from short walks of young and elderly subjects. Results show that instantaneous complexity is able to effectively track the system dynamics and is not affected by statistical noise properties., Massachusetts General Hospital. Dept. of Anesthesia and Critical Care, Harvard Medical School

Autism Spectrum Disorder (ASD) is a neural development disorder characterized by specific patterns of behavioral and social difficulties. Beyond these core symptoms, additional problems such as absence of gender differences identification, interactional distortions of environmental and family responses are often present. Taking into account these emotional and behavioral problems researchers and clinicians are focusing on the design of innovative therapeutic approaches aimed to improve social capabilities of subjects with ASD. Thanks to the technological and scientific progresses of the last years, nowadays it is possible to create human-like robots with social and emotional capabilities. Furthermore it is also possible to analyze physiological signals inferring subjects' psycho-physiological state which can be compared with a behavioral analysis in order to obtain a deeper understanding of subjects reactions to treatments. In this work a preliminary evaluation of an innovative social robot-based treatment for subjects with ASD is described. The treatment consists in a complex stimulation and acquisition platform composed of a social robot, a multi-parametric acquisition system and a therapeutic protocol. During the preliminary tests of the treatment the subject's physiological signals and behavioral parameters have been recorded and used together with the therapists' annotations to infer the subjects' induced reactions. Physiological signals were analyzed and statistically evaluated demonstrating the possibility to correctly discern the two groups (ASD and normally developing subjects) with a classification percentage higher than 92%. Statistical analysis also highlighted the treatment capability to induce different affective states in subjects with ASDs more than in control subjects, demonstrating that the treatment is well designed and tuned on ASDs deficits and behavioral lacks.