Siniša Popović

Physiology-driven adaptive virtual reality stimulation for prevention and treatment of stress related disorders

The significant proportion of severe psychological problems related to intensive stress in recent large peacekeeping operations underscores the importance of effective methods for strengthening the prevention and treatment of stress-related disorders. Adaptive control of virtual reality (VR) stimulation presented in this work, based on estimation of the persons emotional state from physiological signals, may enhance existing stress inoculation training (SIT). Physiology-driven adaptive VR stimulation can tailor the progress of stressful stimuli delivery to the physiological characteristics of each individual, which is indicated for improvement in stress resistance. Following an overview of physiology-driven adaptive VR stimulation, its major functional subsystems are described in more detail. A specific algorithm of stimuli delivery applicable to SIT is outlined.

Comparative analysis of emotion estimation methods based on physiological measurements for real-time applications

In order to improve intelligent Human-Computer Interaction it is important to create a personalized adaptive emotion estimator that is able to learn over time emotional response idiosyncrasies of individual person and thus enhance estimation accuracy. This paper, with the aim of identifying preferable methods for such a concept, presents an experiment-based comparative study of seven feature reduction and seven machine learning methods commonly used for emotion estimation based on physiological signals. The analysis was performed on data obtained in an emotion elicitation experiment involving 14 participants. Specific discrete emotions were targeted with stimuli from the International Affective Picture System database. The experiment was necessary to achieve the uniformity in the various aspects of emotion elicitation, data processing, feature calculation, self-reporting procedures and estimation evaluation, in order to avoid inconsistency problems that arise when results from studies that use different emotion-related databases are mutually compared. The results of the performed experiment indicate that the combination of a multilayer perceptron (MLP) with sequential floating forward selection (SFFS) exhibited the highest accuracy in discrete emotion classification based on physiological features calculated from ECG, respiration, skin conductance and skin temperature. Using leave-one-session-out crossvalidation method, 60.3% accuracy in classification of 5 discrete emotions (sadness, disgust, fear, happiness and neutral) was obtained. In order to identify which methods may be the most suitable for real-time estimator adaptation, execution and learning times of emotion estimators were also comparatively analyzed. Based on this analysis, preferred feature reduction method for real-time estimator adaptation was minimum redundancy - maximum relevance (mRMR), which was the fastest approach in terms of combined execution and learning time, as well as the second best in accuracy, after SFFS. In combination with mRMR, highest accuracies were achieved by k-nearest neighbor (kNN) and MLP with negligible difference (50.33% versus 50.54%); however, mRMR+kNN is preferable option for real-time estimator adaptation due to considerably lower combined execution and learning time of kNN versus MLP.

Real-Time Emotional State Estimator for Adaptive Virtual Reality Stimulation

The paper presents design and evaluation of emotional state estimator based on artificial neural networks for physiology-driven adaptive virtual reality (VR) stimulation. Real-time emotional state estimation from physiological signals enables adapting the stimulations to the emotional response of each individual. Estimation is first evaluated on artificial subjects, which are convenient during software development and testing of physiology-driven adaptive VR stimulation. Artificial subjects are implemented in the form of parameterized skin conductance and heart rate generators that respond to emotional inputs. Emotional inputs are a temporal sequence of valence/arousal annotations, which quantitatively express emotion along unpleasant-pleasant and calm-aroused axes. Preliminary evaluation of emotional state estimation is also performed with a limited set of humans. Human physiological signals are acquired during simultaneous presentation of static pictures and sounds from valence/arousalannotated International Affective Picture System and International Affective Digitized Sounds databases.

A system for head-neck rehabilitation exercises based on serious gaming and virtual reality

Acute and chronic neck pain are common medical conditions, and the treatment typically includes physical therapy involving daily exercises. Insufficient motivation of people afflicted with neck pain to adhere to the prescribed exercise regimen may delay their recovery. Accordingly, in this work, we propose a system that motivates the users to perform neck exercises by engaging them in a serious exergame within virtual reality (VR) environment. The system measures the users’ neck movements via a few static and dynamic kinematic tests and a novel VR serious game, tailored to the neck range of motion of each individual user. The game is designed to make the users perform rehabilitative neck movements according to the prescribed exercise regimen while playing. The analysis of acquired data from VR hardware provides insight into flexibility of the neck during head movements and overall neck kinematics, which is valuable for assessment of pain-related stiffness, as well as for progress monitoring. In a user study performed with the proposed system and the Oculus Rift DK2 VR headset, we show that the users find exercising more interesting and engaging when using the proposed system, and that introducing visually rich VR environments makes the users more motivated to continue exercising.

Interactive scenario control in virtual environments

abstract The important needs of training, exposure, and adoption of specific skills and operating procedures in a safe and cost-effective way strongly encourage the use of virtual environments. A critical component in such environments is interactive control of events and objects that are presented to a trainee. In this paper, we present a platform named the Interactive Scene Control Environment that is composed of trainee and trainer supervision parts. The trainer supervision part enables interactive control of placement and behavior of the objects inserted in the trainees’ scene. The overall system architecture for such a platform is developed and a mechanism for scenario definition is proposed.

fMRI neural activation patterns induced by professional military training

Professional military training makes tough demands on soldiers’ perceptual and motor skills, as well as on their physical fitness and cognitive capabilities in the course of preparation for stressful operational environments. In this pilot study we attempted to identify difference in pattern of neural responses between extensively trained, professional mission-ready soldiers and novice soldiers during audiovisual simulation of mission conditions. We performed fMRI scanning on a few volunteers during presentation of semantically relevant video-clips of real combat from Afghanistan to evaluate influence of military training on mental responses of soldiers. We showed that for professional mission-ready soldiers a week before their deployment to Afghanistan, videoclips with deadly ambush combat induce greater overall brain activation compared to novice soldiers. Missionready soldiers showed greater activation in premotor/prefrontal cortex, posterior parietal cortex, and posterior temporal cortex. These results imply that fMRI technique could be used as challenging step forward in the multidimensional evaluation of military training influence on neural responses and operational capabilities of professional soldiers. This is extremely important not only for potential failure prevention and mere success of the mission, but even more for the survival and the well-being of the servicemen and servicewomen.

Emotionally Based Strategic Communications and Societal Stress-Related Disorders

This article discusses the potential of emotionally based strategic communications (EBSCs) as an extension of traditional strategic communications in prevention of societal stress-related disorders. The concept of EBSCs takes into consideration dominant emotional maps of a specific sociocultural environment in which communications take place. EBSCs may have a significant potential to transform mainly negative-dominant emotional maps of targeted social groups into more positive ones, as a precondition of building a more resilient and stress-resistant social environment. A better understanding of dominant emotional maps and their conditioning may facilitate restoration of more positive emotional maps by touching the right emotions of significant parts of the targeted social groups in the right way. Dominant emotional maps of societies afflicted by economic downturns, natural disasters, conflicts etc., are typically characterized by negatively valenced emotions. Persistent negatively valenced group-based dominant emotions may be used as a quantitative statistical measure of potential stress-related disorders and post-traumatic stress disorders among respected group members. The toxic power of extreme negative emotions, attitudes, actions, and behavior might be reduced by EBSCs as a communication method for transforming negative-dominant emotional maps into more positive ones. EBSCs are conceptualized as the positively valenced stimulation of a negatively emotionally affected group by an appropriate communication strategy to minimize dominant-negative emotional maps and behavior of the targeted group.

A convolutional neural network based approach to QRS detection.

In this paper we present a QRS detection algorithm based on pattern recognition as well as a new approach to ECG baseline wander removal and signal normalization. Each point of the zero-centred and normalized ECG signal is a QRS candidate, while a 1-D CNN classifier serves as a decision rule. Positive outputs from the CNN are clustered to form final QRS detections. The data is obtained from the 44 non-pacemaker recordings of the MIT-BIH arrhythmia database. Classifier was trained on 22 recordings and the remaining ones are used for performance evaluation. Our method achieves a sensitivity of 99.81% and 99.93% positive predictive value, which is comparable with most state-of-the-art solutions. This approach opens new possibilities for improvements in heartbeat classification as well as P and T wave detection problems.