Susana Brás

Mathematical biomarkers for the autonomic regulation of cardiovascular system

Heart rate and blood pressure are the most important vital signs in diagnosing disease. Both heart rate and blood pressure are characterized by a high degree of short term variability from moment to moment, medium term over the normal day and night as well as in the very long term over months to years. The study of new mathematical algorithms to evaluate the variability of these cardiovascular parameters has a high potential in the development of new methods for early detection of cardiovascular disease, to establish differential diagnosis with possible therapeutic consequences. The autonomic nervous system is a major player in the general adaptive reaction to stress and disease. The quantitative prediction of the autonomic interactions in multiple control loops pathways of cardiovascular system is directly applicable to clinical situations. Exploration of new multimodal analytical techniques for the variability of cardiovascular system may detect new approaches for deterministic parameter identification. A multimodal analysis of cardiovascular signals can be studied by evaluating their amplitudes, phases, time domain patterns, and sensitivity to imposed stimuli, i.e., drugs blocking the autonomic system. The causal effects, gains, and dynamic relationships may be studied through dynamical fuzzy logic models, such as the discrete-time model and discrete-event model. We expect an increase in accuracy of modeling and a better estimation of the heart rate and blood pressure time series, which could be of benefit for intelligent patient monitoring. We foresee that identifying quantitative mathematical biomarkers for autonomic nervous system will allow individual therapy adjustments to aim at the most favorable sympathetic-parasympathetic balance.

An automatic classifier of emotions built from entropy of noise

The electrocardiogram (ECG) signal has been widely used to study the physiological substrates of emotion. However, searching for better filtering techniques in order to obtain a signal with better quality and with the maximum relevant information remains an important issue for researchers in this field. Signal processing is largely performed for ECG analysis and interpretation, but this process can be susceptible to error in the delineation phase. In addition, it can lead to the loss of important information that is usually considered as noise and, consequently, discarded from the analysis. The goal of this study was to evaluate if the ECG noise allows for the classification of emotions, while using its entropy as an input in a decision tree classifier. We collected the ECG signal from 25 healthy participants while they were presented with videos eliciting negative (fear and disgust) and neutral emotions. The results indicated that the neutral condition showed a perfect identification (100%), whereas the classification of negative emotions indicated good identification performances (60% of sensitivity and 80% of specificity). These results suggest that the entropy of noise contains relevant information that can be useful to improve the analysis of the physiological correlates of emotion.

ECG delineation and morphological analysis for firefighters tasks differentiation

Between first responder professionals, firefighters registered the highest number of deaths on duty. An abnormal high proportion is associated with cardiovascular events. Our main goal is to identify fatigue/stress during daily routine activities, focusing on the cardiovascular analysis. To accomplish this purpose, ECG wave morphological alterations are analyzed. It was observed that the RR, PP and ST segment significantly differentiate the most stressful tasks from the others.

Associating ECG features with firefighter's activities

In this paper we associate features obtained from ECG signals with the expected levels of stress of real firefighters in action when facing specific events such as fires or car accidents. Five firefighters were monitored using wearable technology collecting ECG signals. Heart rate and heart rate variability features were analyzed in consecutive 5-min intervals during several types of events. A questionnaire was used to rank these types of events according to stress and fatigue and a measure of association was applied to compare this ranking to the ECG features. Results indicate associations between this ranking and both heart rate and heart rate variability features extracted in the time domain. Finally, an example of differences in inter personal responses to stressful events is shown and discussed, motivating future challenges within this research field.

ECG biometric identification: A compression based approach.

Using the electrocardiogram signal (ECG) to identify and/or authenticate persons are problems still lacking satisfactory solutions. Yet, ECG possesses characteristics that are unique or difficult to get from other signals used in biometrics: (1) it requires contact and liveliness for acquisition (2) it changes under stress, rendering it potentially useless if acquired under threatening. Our main objective is to present an innovative and robust solution to the above-mentioned problem. To successfully conduct this goal, we rely on information-theoretic data models for data compression and on similarity metrics related to the approximation of the Kolmogorov complexity. The proposed measure allows the comparison of two (or more) ECG segments, without having to follow traditional approaches that require heartbeat segmentation (described as highly influenced by external or internal interferences). As a first approach, the method was able to cluster the data in three groups: identical record, same participant, different participant, by the stratification of the proposed measure with values near 0 for the same participant and closer to 1 for different participants. A leave-one-out strategy was implemented in order to identify the participant in the database based on his/her ECG. A 1NN classifier was implemented, using as distance measure the method proposed in this work. The classifier was able to identify correctly almost all participants, with an accuracy of 99% in the database used.

Biometric and Emotion Identification: An ECG Compression Based Method

We present an innovative and robust solution to both biometric and emotion identification using the electrocardiogram (ECG). The ECG represents the electrical signal that comes from the contraction of the heart muscles, indirectly representing the flow of blood inside the heart, it is known to convey a key that allows biometric identification. Moreover, due to its relationship with the nervous system, it also varies as a function of the emotional state. The use of information-theoretic data models, associated with data compression algorithms, allowed to effectively compare ECG records and infer the person identity, as well as emotional state at the time of data collection. The proposed method does not require ECG wave delineation or alignment, which reduces preprocessing error. The method is divided into three steps: (1) conversion of the real-valued ECG record into a symbolic time-series, using a quantization process; (2) conditional compression of the symbolic representation of the ECG, using the symbolic ECG records stored in the database as reference; (3) identification of the ECG record class, using a 1-NN (nearest neighbor) classifier. We obtained over 98% of accuracy in biometric identification, whereas in emotion recognition we attained over 90%. Therefore, the method adequately identify the person, and his/her emotion. Also, the proposed method is flexible and may be adapted to different problems, by the alteration of the templates for training the model.

Impact of the Acquisition Time on ECG Compression-Based Biometric Identification Systems

The ECG signal conveys desirable characteristics for biometric identification (universality, uniqueness, measurability, acceptability and circumvention avoidance). However, based on the current literature review, there are no results that evaluate the number of heartbeats needed for personal identification. This information is undoubtedly useful when building a biometric identification system – any system should ask participants to provide data for identification, using the smallest time interval that is possible, for practical reasons. In this paper, we aim at exploring this topic using a measure of similarity based on the Kolmogorov Complexity, called the Normalized Relative Compression (NRC). To attain the goal, we built finite-context models to represent each individual – a compression-based approach that has been shown successful for several other pattern recognition applications like image similarity, DNA sequences or authorship attribution.

BeMonitored: Monitoring psychophysiology and behavior using Android in phobias

It is of the utmost importance that researchers can recreate, as accurately as possible, real-life conditions in psychological studies. However, that is not always possible. Given that phobias are rather context-specific, their study is the ideal candidate to assess the feasibility of using a mobile and wearable device for obtaining physiological and behavioral data. In this article, we propose BeMonitored, a smartphone-based solution to support more ecologically valid monitoring of psychological experiments. BeMonitored delivers customizable, specific context-dependent audiovisual stimuli and uses external resources connected via Bluetooth or a smartphone’s own resources, while capturing the participant’s behavior, physiology, and environment. We used BeMonitored in a spider phobia case study and showed that spider phobics differed from control participants in face motion, captured by the smartphone camera. Moreover, our results also revealed heart rate differences between spider and neutral stimuli in phobic participants. The presented results emphasize the usefulness of smartphones for phobia monitoring. Considering their intrinsic characteristics, smartphones may constitute the natural evolution from the lab to more realistic contexts.

Psychophysiology of disgust: ECG noise entropy as a biomarker.

The identification or classification of emotions allows the description of the persons state and, therefore, the inference of their preferences. The basic emotion of disgust, in particular, allows the organism to protect itself against diseases. Usually, the decrease in heart rate is associated with this emotion. As an avoidance behavior, when facing with disgust stimuli, the body reacts with movements, such as muscle contraction, etc. These reactions are evidenced in the electrocardiogram (ECG) as noise responses. In this paper, we propose the amount of ECG noise measured by the noise entropy as a new biomarker in emotion identification, which has been neglected in the literature. Our results showed that the noise entropy was able to discriminate between disgust, fear and neutral conditions in 88% (p<;0.05). It was also evidenced in this dataset that the median noise entropy in disgust was higher than in neutral and in fear conditions.

Monitoring physiology and behavior using Android in phobias.

In this paper, we present an Android-based system Application - AWARE - for the assessment of the persons physiology and behavior outside of the laboratory. To accomplish this purpose, AWARE delivers context dependent audio-visual stimuli, embedded into the subjects real-world perception, via marker/vision-based augmented reality (AR) technology. In addition, it employs external measuring resources connected via Bluetooth, as well as the smartphones integrated resources. It synchronously acquires the experiments video (camera input with AR overlay), physiologic responses (with a dedicated ECG measuring device) and behavior (through movement and location, with accelerometer/gyroscope and GPS, respectively). Psychological assessment is heavily based on laboratory procedures, even though it is known that these settings disturb the subjects natural reactions and condition. The major idea of this application is to evaluate the participant condition, mimicking his/her real life conditions. Given that phobias are rather context specific, they represent the ideal candidate for assessing the feasibility of a mobile system application. AWARE allowed presenting AR stimuli (e.g., 3D spiders) and quantifying the subjects reactions non-intrusively (e.g., heart rate variation) - more emphatic in the phobic volunteer when presented with spider vs non phobic stimulus. Although still a proof of concept, AWARE proved to be flexible, and straightforward to setup, with the potential to support ecologically valid monitoring experiments.