Carlos Carreiras

Outlier Detection in Non-intrusive ECG Biometric System

A recent trend in the field of biometrics is the use of Electrocardiographic (ECG) signals. One of the main challenges of this new paradigm is the development of non-intrusive and highly usable setups. Fingers and hand palms for example, allow the ECG acquisition at much more convenient locations than the chest, commonly used for clinical scenarios. These new locations lead to an ECG signal with lower signal to noise ratio, and more prone to noise artifacts. In this paper, we propose an outlier removal system to eliminate noisy segments, and enhance the performance of non-intrusive ECG biometric systems. Preliminary results show that this system leads to an improvement on the recognition rates, helping to further validate the potential of ECG signals as complementary biometric modality.

A web-based platform for biosignal visualization and annotation

With the advent of wearable sensing and mobile technologies, biosignals have seen an increasingly growing number of application areas, leading to the collection of large volumes of data. One of the difficulties in dealing with these data sets, and in the development of automated machine learning systems which use them as input, is the lack of reliable ground truth information. In this paper we present a new web-based platform for visualization, retrieval and annotation of biosignals by non-technical users, aimed at improving the process of ground truth collection for biomedical applications. Moreover, a novel extendable and scalable data representation model and persistency framework is presented. The results of the experimental evaluation with possible users has further confirmed the potential of the presented framework.

A Unifying Approach to ECG Biometric Recognition Using the Wavelet Transform

Biometric recognition systems use measures from the body itself to determine the identity of an individual. The electrocardiogram (ECG) has been increasingly used as a biometric measure for person identification, as it is an easily measurable characteristic of all individuals. Our method for ECG acquisition follows an off-the-person approach, using a single ECG lead with non-gelled electrodes placed at the hands. However, this signal is noisier than typical ECG signals acquired on the chest, making subsequent processing more difficult. Therefore, we investigate the applicability of the Wavelet Transform (WT), which decomposes a signal into a time-scale representation according to a given mother wavelet. We use this representation to both segment the R wave of the ECG signal, and as the features for the classification step, defining an appropriate distance measure. We test this framework with real data, using various mother wavelets. Our experimental results show the potential of this framework, and that the best mother wavelet for the evaluated context is the rbio5.5.

Towards a Continuous Biometric System Based on ECG Signals Acquired on the Steering Wheel

Electrocardiogram signals acquired through a steering wheel could be the key to seamless, highly comfortable, and continuous human recognition in driving settings. This paper focuses on the enhancement of the unprecedented lesser quality of such signals, through the combination of Savitzky-Golay and moving average filters, followed by outlier detection and removal based on normalised cross-correlation and clustering, which was able to render ensemble heartbeats of significantly higher quality. Discrete Cosine Transform (DCT) and Haar transform features were extracted and fed to decision methods based on Support Vector Machines (SVM), k-Nearest Neighbours (kNN), Multilayer Perceptrons (MLP), and Gaussian Mixture Models - Universal Background Models (GMM-UBM) classifiers, for both identification and authentication tasks. Additional techniques of user-tuned authentication and past score weighting were also studied. The method’s performance was comparable to some of the best recent state-of-the-art methods (94.9% identification rate (IDR) and 2.66% authentication equal error rate (EER)), despite lesser results with scarce train data (70.9% IDR and 11.8% EER). It was concluded that the method was suitable for biometric recognition with driving electrocardiogram signals, and could, with future developments, be used on a continuous system in seamless and highly noisy settings.

ECG biometrics: A template selection approach

Electrocardiography (ECG) biometrics is emerging as a viable biometric trait. Recent developments at the sensor level have shown the feasibility of performing signal acquisition at the fingers and hand palms, using one-lead sensor technology and dry electrodes. These new locations lead to ECG signals with lower signal to noise ratio and more prone to noise artifacts; the heart rate variability is another of the major challenges of this biometric trait. In this paper we propose a novel approach to ECG biometrics, with the purpose of reducing the computational complexity and increasing the robustness of the recognition process enabling the fusion of information across sessions. Our approach is based on clustering, grouping individual heartbeats based on their morphology. We study several methods to perform automatic template selection and account for variations observed in a persons biometric data. This approach allows the identification of different template groupings, taking into account the heart rate variability, and the removal of outliers due to noise artifacts. Experimental evaluation on real world data demonstrates the advantages of our approach.

ECG signals for biometric applications - are we there yet?

The potential of the electrocardiographic (ECG) signal as a biometric trait has been ascertained in the literature over the past decade. The inherent characteristics of the ECG make it an interesting biometric modality, given its universality, intrinsic aliveness detection, continuous availability, and inbuilt hidden nature. These properties enable the development of novel applications, where non-intrusive and continuous authentication are critical factors. Examples include, among others, electronic trading platforms, the gaming industry, and the auto industry, in particular for car sharing programs and fleet management solutions. However, there are still some challenges to overcome in order to make the ECG a widely accepted biometric. In particular, the questions of uniqueness (inter-subject variability) and permanence over time (intra-subject variability) are still largely unanswered. In this paper we focus on the uniqueness question, presenting a preliminary study of our biometric recognition system, testing it on a database encompassing 618 subjects. We also performed tests with subsets of this population. The results reinforce that the ECG is a viable trait for biometrics, having obtained an Equal Error Rate of 9.01% and an Error of Identification of 15.64% for the entire test population.

Unsupervised Analysis of Morphological ECG Features for Attention Detection

Physiological Computing augments the information bandwidth between a computer and its user by continuous, real-time monitoring of the user’s physiological traits and responses. This is especially interesting in a context of emotional assessment during human-computer interaction. The electroencephalogram (EEG) signal, acquired on the scalp, has been extensively used to understand cognitive function, and in particular emotion. However, this type of signal has several drawbacks, being susceptible to noise and requiring the use of impractical head-mounted apparatuses. For these reasons, the electrocardiogram (ECG) has been proposed as an alternative source to assess emotion, which is continuously available, and related with the psychophysiological state of the subject. In this paper we analyze morphological features of the ECG signal acquired from subjects performing an attention-demanding task. The analysis is based on various unsupervised learning techniques, which are validated against evidence found in a previous study by our team, where EEG signals collected for the same task exhibit distinct patterns as the subjects progress in the task.

CardioWheel: ECG Biometrics on the Steering Wheel

Monitoring physiological signals while driving is a recent trend in the automotive industry. We present CardioWheel, a state-of-the-art machine learning solution for driver biometrics based on electrocardiographic signals ECG. The presented system pervasively acquires heart signals from the users hands through sensors embedded in the steering wheel, to recognize the drivers identity. It combines unsupervised and supervised machine learning algorithms, and is being tested in real-world scenarios, illustrating one of the potential uses of this technology.

Dominant Set Approach to ECG Biometrics

Electrocardiographic ECG signals are emerging as a recent trend in the field of biometrics. In this paper, we propose a novel ECG biometric system that combines clustering and classification methodologies. Our approach is based on dominant-set clustering, and provides a framework for outlier removal and template selection. It enhances the typical workflows, by making them better suited to new ECG acquisition paradigms that use fingers or hand palms, which lead to signals with lower signal to noise ratio, and more prone to noise artifacts. Preliminary results show the potential of the approach, helping to further validate the highly usable setups and ECG signals as a complementary biometric modality.

Evaluating Template Uniqueness in ECG Biometrics

Research over the past decade has demonstrated the capability of the electrocardiographic (ECG) signal to be used as a biometric trait, through which the identity of an individual can be recognized. Given its universality, intrinsic aliveness detection, continuous availability, and inherent hidden nature, the ECG is an interesting biometric modality enabling the development of novel applications, where non-intrusive and continuous authentication are critical factors. Examples include personal computers, the gaming industry, and the auto industry, especially for car sharing programs and fleet management solutions. Nonetheless, from a theoretical point of view, there are still some challenges to overcome in bringing ECG biometrics to mass markets. In particular, the issues of uniqueness (related to inter-subject variability) and permanence (related to intra-subject variability) are still largely unanswered. This work focuses on the uniqueness issue, evaluating the performance of our ECG biometric system over a database encompassing 618 subjects. Additionally, we performed tests with subsets of this population. The results cement the ECG as a viable trait to be used for identity recognition, having obtained and Equal Error Rate of \(9.01\,\%\) and an Error of Identification of \(15.64\,\%\) for the entire test population.