BIOSIGNALS 2011 Abstracts


Full Papers
Paper Nr: 11
Title:

COMPLEXITY ANALYSIS OF MASS SPECTROMETRY DATA FOR DISEASE CLASSIFICATION USING GA-BASED MULTISCALE ENTROPY

Authors:

Cuong C. To and Tuan D. Pham

Abstract: Entropy methods including approximate entropy (ApEn), sample entropy (SampEn) and multiscale entropy (MSE) have recently been applied to measure the complexity of finite length time series for classification of diseases. In order to effectively use these entropy methods, parameters such as m, r, and scale factor (in MSE) are to be determined. So far, there have been no general rules to select these parameters as they depend on particular problems. In this paper, we introduce a genetic algorithm (GA) based method for optimal selection of these parameters in a sense that the entropic difference between healthy and pathologic groups are maximized.
Download

Paper Nr: 21
Title:

AUTOMATED BURST DETECTION IN NEONATAL EEG

Authors:

Sourya Bhattacharyya, Jayanta Mukhopadhyay, Arun Kumar Majumdar, Bandana Majumdar, Arun Kumar Singh and Chanchal Saha

Abstract: Presence of burst suppression pattern in neonate EEG is a sign of epilepsy. Detection of burst patterns is normally done by visual inspection of recorded raw EEG or amplitude integrated EEG signal. Existing automatic burst detection approaches consist of either supervised learning mechanism or static energy threshold based comparison. Both approaches can produce inconsistent results for babies with different ages (for example, a neonate EEG and a six month old baby EEG). That is because, EEG signal amplitude or energy increases according to baby’s age. Training based classifiers or static thresholds cannot adapt with this amplitude variation. Here we propose an automatic burst detection method, which first computes signal parameters such as energy, variance and power spectral density. From generated signal data, so called low level amplitude or energy output is used as a ground reference for indication of signal suppression level. Burst is identified according to high deviation of parameter values from those in suppression pattern. It does not need any static threshold based comparison. Results show that our algorithm exhibits greater sensitivity and equal specificity than existing methods. Due to adaptive thresholding for burst detection, our method is applicable for analyzing EEG signals of babies with different ages.
Download

Paper Nr: 25
Title:

OPTIMAL CONTROL OF MIXED-STATE QUANTUM SYSTEMS BASED ON LYAPUNOV METHOD

Authors:

Yuanyuan Zhang, Shuang Cong and Kezhi Li

Abstract: An optimal control strategy of mixed state steering in finite-dimensional closed quantum systems is proposed in this paper. Two different situations are considered: one is the target state is in statistical incoherent mixtures of energy eigenstates in which the target states are diagonal. Another is not all of the off-diagonal elements in the target states are zeros. We change the trajectory tracking problem into the state steering one by introducing the unitary transformation with all energy eigenstates in the inner Hamiltonian of system controlled. Based on Lyapunov stability theorem the stable parameters of controller designed is selected and the optimality of the control law proposed is proven. Moreover, two numerical system control simulations are performed on the diatomic molecule described by the Morse oscillator model under the control law proposed. The system control simulation experimental results demonstrate that the control strategies proposed are efficient even when the controlled system is not completely controllable.
Download

Paper Nr: 28
Title:

IMPROVEMENT AND VALIDATION OF AN AUTOMATED NEONATAL SEIZURE DETECTOR

Authors:

P. J. Cherian, W. Deburchgraeve, V. Matic, M. De Vos, R. M. Swarte, J. H. Blok, P. Govaert, S. Van Huffel and G. H. Visser

Abstract: We present the improvements made to and subsequent validation of an automated approach to detect neonatal seizures. The evaluation of the algorithm has been performed on a new and extensive data set of neonatal EEGs. Previously, we have classified neonatal seizures visually into two types: the spike train and oscillatory type of seizures and developed two separate algorithms that run in parallel for their automated detection. The first algorithm analyzes the correlation between high-energetic segments of the EEG, whereas the second one detects increases in low-frequency activity (<8 Hz) and then uses an autocorrelation. An improved version of our automated system (called ‘NeoGuard’) uses more informative features for classification and optimized parameters for thresholding. The validation was performed on 756 hours of ‘unseen’ continuous EEG monitoring data from 24 neonates with encephalopathy and recorded seizures. The seizure detection system showed a median sensitivity of 86.9 % per patient, positive predictive value (PPV) of 89.5 % and false positive rate of 0.28 per hour. The modified algorithm has a high sensitivity combined with a good PPV whereas false positive rate is much lower compared to the previous version of the algorithm.
Download

Paper Nr: 45
Title:

A PROPOSAL OF A NOVEL CARDIORESPIRATORY LONG-TERM MONITORING DEVICE

Authors:

S. Lapi, E. Biagi, G. Borgioli, M. Calzolai, L. Masotti and G. Fontana

Abstract: Monitoring of respiratory movements is an important feature in planning of medical care. We present here a simple, portable, accelerometer-based device suitable for long term-monitoring of the breathing and heart rates, along with postural changes, during sleep and wakefulness. Recordings of respiratory frequency, heart rate, posture and voluntary cough were obtained from a group of volunteers who also participated in sleep studies (6-8 hrs). A pair of capacitive MEMS tri-axial accelerometers was positioned at the level of the 10th rib along the mid-axillary line bilaterally; simultaneous recordings of respiratory movements, heart rate and body position could be easily performed. The signal were digitized and used to detect body position and relative movement between accelerometers. Conventional spirometry was performed in parallel when appropriate. During resting breathing, qualitative analysis revealed that the accelerometric assessment of respiratory pattern correlated well with that obtained by spirometry. Values of respiratory rates were identical with the two techniques. Recordings of respiratory and cardiac activity during sleep were satisfactorily obtained except for short lasting episodes corresponding to changes in body position. These devices seem to be also suitable for detecting the motor pattern of cough.
Download

Paper Nr: 48
Title:

NEURAL-FUZZY MODELLING OF LUNG VOLUME USING ABSOLUTE ELECTRICAL IMPEDANCE TOMOGRAPHY

Authors:

Suzani Mohamad Samuri, George Panoutsos, Mahdi Mahfouf, G. H. Mills, M. Denaï and B. H. Brown

Abstract: Electrical Impedance Tomography (EIT) has been the subject of intensive research since its development in the early 1980s by Barber and Brown at the Department of Medical Physics and Clinical Engineering, Hallamshire Hospital in Sheffield (UK). In particular, pulmonary measurement has been the focus of most EIT related research. One of the relatively recent advances in EIT is the development of an absolute EIT system (aEIT) which can estimate absolute values of lung resistivity and lung volumes. However, there is still active research in the area of validating and improving the accuracy and consistency of the aEIT estimation of lung volumes towards characterising the system as suitable for clinical use. In this paper we present a new approach based on Computational Intelligence (CI) modelling to model the ‘Resistivity - Lung Volume’ relationship that will allow more accurate lung volume predictions. Eight (8) healthy volunteers were measured simultaneously by the Sheffield aEIT system and a Spirometer and the recorded results were used to develop subject-specific Neural-Fuzzy models able to predict absolute values of lung volume based only on absolute lung resistivity data. The developed models show improved accuracy in the prediction of lung volumes, as compared with the original Sheffield aEIT system. However the inter-individual differences observed in the subject-specific modelling behaviour of the ‘Resistivity-Lung Volume’ curves suggest that a model extension is needed, whereby the modelling structure auto-calibrates to account for subject (or patient-specific) inter-parameter variability.
Download

Paper Nr: 59
Title:

BIO-INSPIRED AUDITORY PROCESSING FOR SPEECH FEATURE ENHANCEMENT

Authors:

HariKrishna Maganti and Marco Matassoni

Abstract: Mel-frequency cepstrum based features have been traditionally used for speech recognition in a number of applications, as they naturally provide a higher recognition accuracies. However, these features are not very robust in a noisy acoustic conditions. In this article, we investigate the use of bio-inspired auditory features emulating the processing performed by cochlea to improve the robustness, particularly to counter environmental reverberation. Our methodology first extracts robust noise resistant features by gammatone filtering, which emulate cochlea frequency resolution and then a long-term modulation spectral processing is performed which preserves speech intelligibility in the signal. We compare and discuss the features based upon the performance on Aurora5 meeting recorder digit task recorded with four different microphones in a hands-free mode at a real meeting room. The experimental results show that the proposed features provide considerable improvements with respect to the state of the art feature extraction techniques.
Download

Paper Nr: 76
Title:

INFERENCE OF BRAIN MENTAL STATES FROM SPATIO-TEMPORAL ANALYSIS OF EEG SINGLE TRIALS

Authors:

Yehudit Hasson-Meir, Andrey Zhdanov, Talma Hendler and Nathan Intrator

Abstract: We present an efficient and robust computational model for brain state interpretation from EEG single trials. This includes identification of the most relevant time points and electrodes that may be active and contribute to differentiation between the mental states investigated during the experiment. The model includes a regularized logistic regression classifier trained with cross-validation to find the optimal model and its regularization parameter. The proposed framework is generic and can be applied to different classification tasks. In this study we applied it to a classical visual task of distinction between faces and houses. The results show that the obtained single trial prediction is significantly better than chance. Moreover, correct choice of the regularization parameter significantly improves classification results. In addition, the obtained spatial-temporal information of brain activity can give an indication to correlated activity of regions of the brain (spatial) as well as temporal activity correlations between and within EEG electrodes. This spatial-temporal analysis can render a far more holistic interpretability for visual perception mechanism without any a priori bias on certain time periods or scalp locations.
Download

Paper Nr: 89
Title:

FEATURE SELECTION FOR INTER-PATIENT SUPERVISED HEART BEAT CLASSIFICATION

Authors:

G. Doquire, G. de Lannoy, D. François and M. Verleysen

Abstract: Supervised and inter-patient classification of heart beats is primordial in many applications requiring long-term monitoring of the cardiac function. Several classification models able to cope with the strong class unbalance and a large variety of ECG feature sets have been proposed for this task. In practice, over 200 features are often considered and the features retained in the final model are either chosen using domain knowledge or an exhaustive search in the feature sets without evaluating the relevance of each individual feature included in the classifier. As a consequence, the results obtained by these models can be suboptimal and difficult to interpret. In this work, feature selection techniques are considered to extract optimal feature subsets for state of the art ECG classification models. The performances are evaluated on real ambulatory recordings and compared to previously reported feature choices using the same models. Results indicate that a small number of individual features actually serve the classification and that better performances can be achieved by removing useless features.
Download

Paper Nr: 98
Title:

OPTICAL METHODS FOR LOCAL PULSE WAVE VELOCITY ASSESSMENT

Authors:

T. Pereira, M. Cabeleira, P. Matos, E. Borges, V. Almeida, J. Cardoso, C. Correia and H. C. Pereira

Abstract: Pulse wave velocity (PWV) is a clinically interesting parameter associated to cardiac risk due to arterial stiffness, generally evaluated by the time that the pressure wave spends to travel between two arbitrary points. Optic sensors are an attractive instrumental solution in this kind of time assessment applications due to their truly non-contact operation capability, which ensures an interference free measurement. On the other hand, they can pose different challenges to the designer, mostly related to the features of the signals they produce and to the associated signal processing burden required to extract error free, reliable information. In this work we evaluate two prototype optical probes dedicated to pulse transit time (PTT) evaluation as well as three algorithms for its assessment. Although the tests were carried out at the test bench, where “well behaved” signals can be obtained, the transition to a probe for use in humans is also considered. Results demonstrated the possibility of measuring pulse transit times as short as 1 ms with less than 1% error.
Download

Paper Nr: 164
Title:

A REAL-TIME FRACTAL-BASED BRAIN STATE RECOGNITION FROM EEG AND ITS APPLICATIONS

Authors:

Olga Sourina, Qiang Wang, Yisi Liu and Minh Khoa Nguyen

Abstract: EEG-based immersion is a new direction in research and development on human computer interfaces. It has attracted recently more attention from the research community and industry as wireless EEG reading devices became easily available on the market. EEG-based technology has been applied in anaesthesiology, psychology, serious games or even in marketing. As EEG signal is considered to have a fractal nature, we proposed and developed a novel spatio-temporal fractal based approach to the brain state quantification. The real-time algorithms of emotion recognition and concentration level recognition were implemented and integrated in human-computer interfaces of EEG-enable applications. In this paper, EEG-based “serious” games for concentration training and emotion-enable applications including emotion-based music therapy on the Web were proposed and implemented.
Download

Paper Nr: 166
Title:

UNDERSTANDING CEREBRAL ACTIVATIONS IN NEUROMARKETING - A Neuroelectrical Perspective

Authors:

Giovanni Vecchiato, Laura Astolfi, Fabrizio De Vico Fallani, Jlenia Toppi, Fabio Aloise, Febo Cincotti, Donatella Mattia and Fabio Babiloni

Abstract: This paper aims to be a survey of recent experiments performed in the Neuromarketing field. Our purpose is to illustrate results obtained by employing the popular tools of investigation well known in the international neuroelectrical community such as the MEG, High Resolution EEG techniques and steady-state visually evoked potentials. By means of temporal and frequency patterns of cortical activations we intend to show how the neuroscientific community is nowadays sensible to the needs of companies and, at the same time, how the same tools are able to retrieve hidden information about the demands of consumers. These instruments could be of help both in pre- and post-design stage of a product, or a service, that a marketer is going to promote.
Download

Short Papers
Paper Nr: 13
Title:

SYNCHRONISATION OF BIOLOGICAL CLOCK SIGNALS - Capturing Coupled Repressilators from a Control Systems Perspective

Authors:

Thomas Hinze, Mathias Schumann and Stefan Schuster

Abstract: Exploration of chronobiological systems emerges as a growing research field within bioinformatics focusing on various applications in medicine, agriculture, and material sciences. From a systems biological perspective, the question arises whether biological control systems for regulation of oscillative signals and their technical counterparts utilise similar mechanisms. If so, modelling approaches and parameterisation adopted from building blocks can help to identify general components for clock synchronisation. Phase-locked loops could be an interesting candidate in this context. Both, biology and engineering, can benefit from a unified view. In a first experimental study, we analyse a model of coupled repressilators. We demonstrate its ability to synchronise clock signals in a monofrequential manner. Several oscillators initially deviate in phase difference and frequency with respect to explicit reaction and diffusion rates. Accordingly, the duration of the synchronisation process depends on dedicated reaction and diffusion parameters whose settings still lack to be sufficiently captured by comprehensive tools like the Kuramoto approach.
Download

Paper Nr: 17
Title:

CONTINUOUS ANALYSIS OF REPOLARIZATION CHARACTERISTICS DURING INSULIN INDUCED HYPOGLYCEMIA

Authors:

J. A. Lipponen, P. A. Karjalainen, M. P. Tarvainen, J. Kemppainen, H. Mikkola, T. Kärki and T. Laitinen

Abstract: Hypoglycemia has been shown to affect ECG. Reported changes are prolongation of QT-interval and increased R/T amplitude ratio. These ECG changes are suggested to be connected to so-called dead in bed syndrome. Continuous analysis of ECG changes and blood glucose values, during insulin induced hypoglycemia is presented. Altogether 22 subjects were analyzed in three different groups; 1) healthy group 2) diabetic patients diagnosed less 5 years ago and 3) chronic diabetics diagnosed over than 5 years ago. The results showed that 20 of 22 subjects’ QT-time was prolonged during hypoglycemia. In addition, in group 3 changes were smaller than in groups 1 and 2.
Download

Paper Nr: 19
Title:

A COMPUTATIONAL MODEL FOR CONSCIOUS VISUAL PERCEPTION AND FIGURE/GROUND SEPARATION

Authors:

Marc Ebner and Stuart Hameroff

Abstract: The human brain is able to perform a number feats that researchers have not been able to replicate in artificial systems. Unsolved questions include: Why are we conscious and how do we process visual information from the input stimulus right down to the individual action. We have created a computational model of visual information processing. A network of spiking neurons, a single layer, is simulated. This layer processes visual information from a virtual retina. In contrast to the standard integrate and fire behavior of biological neurons, we focus on lateral connections between neurons of the same layer. We assume that neurons performing the same function are laterally connected through gap junctions. These lateral connections allow the neurons responding to the same stimulus to synchronize their firing behavior. The lateral connections also enable the neurons to perform figure/ground separation. Even though we describe our model in the context of visual information processing, it is clear that the methods described, can be applied to other kinds of information, e.g. auditory.
Download

Paper Nr: 20
Title:

EFFECTIVE AND ACCELERATED INFORMATIVE FRAME FILTERING IN COLONOSCOPY VIDEOS USING GRAPHICS PROCESSING UNIT

Authors:

Venkata Praveen Karri, JungHwan Oh, Wallapak Tavanapong, Johnny Wong and Piet C. de Groen

Abstract: Colonoscopy is an endoscopic technique that allows a physician to inspect the mucosa of the human colon. It has contributed to a marked decline in the number of colorectal cancer related deaths. However, recent data suggest that there is a significant (4-12%) miss-rate for the detection of even large polyps and cancers. To address this, we have investigated automated post-procedure and real-time quality measurements by analyzing colonoscopy videos. One of the fundamental steps is separating informative frames from non-informative frames, a process called Informative Frame Filtering (IFF). Non-informative frames comprise out-of-focus frames and frames lacking typical features of the colon. We introduce a new IFF algorithm in this paper, which is much more accurate than our previous one. Also, we exploit the many-core GPU (Graphics Processing Unit) to create an IFF software module for High Performance Computing (HPC). Code optimizations embedded in the many-core GPU resulted in a 40-fold acceleration compared to CPU-only implementation for our IFF software module.
Download

Paper Nr: 30
Title:

WEIGHTED TIME WARPING FOR TEMPORAL SEGMENTATION OF MULTI-PARAMETER PHYSIOLOGICAL SIGNALS

Authors:

Gartheeban Ganeshapillai and John Guttag

Abstract: We present a novel approach to segmenting a quasiperiodic multi-parameter physiological signal in the presence of noise and transient corruption. We use Weighted Time Warping (WTW), to combine the partially correlated signals. We then use the relationship between the channels and the repetitive morphology of the time series to partition it into quasiperiodic units by matching it against a constantly evolving template. The method can accurately segment a multi-parameter signal, even when all the individual channels are so corrupted that they cannot be individually segmented. Experiments carried out on MIMIC, a multi-parameter physiological dataset recorded on ICU patients, demonstrate the effectiveness of the method. Our method performs as well as a widely used QRS detector on clean raw data, and outperforms it on corrupted data. Under additive noise at SNR 0 dB the average errors were 5:81 ms for our method and 303:48 ms for the QRS detector. Under transient corruption they were 2:89 ms and 387:32 ms respectively.
Download

Paper Nr: 36
Title:

SLEEPIC - Developments for a Wearable On-line Sleep and Wake Discrimination System

Authors:

Walter Karlen and Dario Floreano

Abstract: The design of wearable systems comes with constraints in computational and power resources. We describe the development of customized hardware for the wearable discrimination of human sleep and wake based on cardio-respiratory signals. The device was designed for efficient and low-power computation of Fast Fourier Transforms and artificial neural networks required for the on-line classification. We discuss methods for reducing computational load and consequently power requirements. The SleePic prototype was tested for autonomy and comfort on eight healthy subjects. SleePic showed an energetic autonomy of more than 36 hours. The SleePic device will require further integration for increased comfort and improved user interaction.
Download

Paper Nr: 38
Title:

ACCELEROMETER BASED GAIT ANALYSIS - Multi Variate Assessment of Fall Risk with FD-NEAT

Authors:

Bart Jansen, Maxine Tan, Ivan Bautmans, Bart Van Keymolen, Tony Mets and Rudi Deklerck

Abstract: This paper describes an accelerometer based gait analysis system for the assessment of fall risk. The assessment is based on 22 different features calculated from the signal. The different features are combined using machine learning algorithms in order to decide whether the subject has an increased fall risk. Results from Naive Bayes, Neural Networks, Locally Weighted Learning, Support Vector Machines and C4.5 are reported and compared. It is argued that the neural networks provide low accuracy results because of the high dimensionality of the feature space compared to the available data. It is shown that FD-NEAT (a method from neuro evolution which simultaneously learns the network topology, the network weights and the relevant features) outperforms the other methods in the given classification task. The system is evaluated on a database consisting of 40 elderly with known fall risk and 40 healthy elderly controls.
Download

Paper Nr: 39
Title:

INVESTIGATION OF THE NON-MARKOVITY SPECTRUM AS A COGNITIVE PROCESSING MEASURE OF DEEP BRAIN MICROELECTRODE RECORDINGS

Authors:

P. A. Meehan, P. A. Bellette, A. P. Bradley, J. E. Castner, H. J. Chenery, D. A. Copland, J. D. Varghese, T. Coyne and P. A. Silburn

Abstract: Previous research has shown that changes in complexity-based measures of deep brain (DB) microelectrode recordings (MER) from conscious human patients, show correlations with different linguistic tasks. These statistical mechanics based measures are further expanded in this research to look at the spectra of an adapted non-Markovity parameter in different frequency ranges as a measure of synchronous neuronal networked behaviour. Results presented show statistically significant interaction between hemisphere of recording, epoch of brain function and semantic category in the fast frequency range (80-200Hz). Processing of similar semantic words appeared to be associated with increased synchrony in the left hand hemisphere. Evidence for substantial left and right hemispherical interactions was found. Similar, but less important trends were found in the beta band (10-30Hz). Significant but less specific correlations were also found in the theta (4-10Hz) and gamma (30-80Hz) frequency bands.
Download

Paper Nr: 42
Title:

A FRAMEWORK FOR ACOUSTIC CARDIAC SIGNAL ANALYSIS

Authors:

P. Carvalho, R. P. Paiva, D. Kumar, J. Ramos, S. Santos and J. Henriques

Abstract: Cardiac auscultation is a traditional, yet highly sensitive and specific diagnosis technique for cardiovascular diseases. We present a Matlab framework for cardiac signals processing and analysis, which includes a new toolbox specifically designed for the main processing tasks related to heart sound analysis. Existing frameworks for acoustic cardiac signal analysis usually limit themselves to noise contamination detection, S1 and S2 segmentation and murmur diagnosis. Besides these operations, the proposed framework includes algorithms developed for segmentation of the main heart sound components capable of handling situations with high-grade murmur, S3 detection and identification, S2 split identification as well as systolic time intervals (STI) measurement using heart sound. Methods for cardiac function parameter extraction based on STI are also included. Most of the algorithms outlined in the paper have been extensively evaluated using data collected from patients with several types of cardiovascular diseases under real-life conditions. The achieved results suggest that the algorithms developed for the framework exhibit performances that are comparable and, in most cases, surpass existing state of the art methods.
Download

Paper Nr: 43
Title:

ROBUST CHARACTERISTIC POINTS FOR ICG - Definition and Comparative Analysis

Authors:

P. Carvalho, R. P. Paiva, J. Henriques, M. Antunes, I. Quintal and J. Muehlsteff

Abstract: The impedance cardiogram (ICG) is a promising tool for non-invasive and cost effective assessment of the hemodynamic state, especially in low acuity and home settings. Important diagnostic parameters are related to characteristic points within the ICG, i.e. the B and the X points that are assumed to mark the opening and closure of the aortic valve., respectively. Based on synchronized echocardiography-ICG data obtained from healthy subjects at rest, we compare 4 existing alternatives for the ICG’s characteristic point definitions associated detection algorithms. We show that those points exhibit considerable biases with respect to the intended onsets of the systole and diastole compared to the Echocardiography Goldstandard. We introduce a new approach to determine these characteristic points based on the analysis of the ICG morphology. For its implementation a computationally simple algorithm, based on high order derivatives, is proposed. This algorithm is evaluated using simultaneously recorded echocardiographies and ICG signals. The achieved results show that the proposed method enables the identification of the main characteristic points, B and X, with significantly smaller errors and much higher correlations compared to current state of the art methods and existing alternative characteristic point definitions.
Download

Paper Nr: 51
Title:

AN IMPROVED APPROACH FOR REAL-TIME DETECTION OF SLEEP APNEA

Authors:

Baile Xie, Wenxun Qiu, Hlaing Minn, Lakshman Tamil and Mehrdad Nourani

Abstract: The traditional diagnosis of sleep apnea and hypopnea syndrome (SAHS) requires an expensive and complex overnight procedure called polysomnography (PSG). Recently, finding valid alternatives for SAHS diagnosis has attracted much research attention. This paper focuses on the real-time monitoring and detection of SAHS based on the arterial oxygen saturation signal measured by pulse oximetry (SpO2). We develop a more comprehensive feature set and a more appropriate annotation criterion, if compared to the existing approaches in the literature. To enjoy competitiveness in computational complexity, we also propose a reduced feature set which provides a higher sensitivity and better adaptivity to distinct databases. The performances of 15 commonly used classifiers with different cost matrixes are assessed on different databases, offering detailed insights on the diagnostic abilities of these methods.
Download

Paper Nr: 54
Title:

DAMPING FACTOR CONSTRAINTS AND METABOLITE PROFILE SELECTION INFLUENCE MAGNETIC RESONANCE SPECTROSCOPY DATA QUANTIFICATION

Authors:

M. I. Osorio Garcia, D. M. Sima, S. Van Huffel, F. U. Nielsen and U. Himmelreich

Abstract: Magnetic Resonance Spectroscopy (MRS) is a technique used for the diagnostics of tumour and metabolic diseases by estimating the metabolite concentrations of the tissue under investigation. Unreliable metabolite estimation may mislead the diagnosis and therefore quantification of MRS in vivo signals must be performed carefully. In this work, we quantify 1.5 Tesla (T) and 9.4 T MRS in vivo signals and study the influence of the damping factor constraint and the metabolite profile selection used in the quantification method. The damping factor bounds the linewidth of the metabolite profiles and may yield bad fits if wrongly selected. Furthermore, MRS data quantification leads to overestimation of some metabolite concentrations when the selected metabolite basis set is incomplete suggesting that metabolites are fitting the region of their neighboring components. Here, we evaluate the normality of the residual which in cases of good fitting contains no metabolites and only white Gaussian noise. Furthermore, we propose to estimate the damping bound adaptively by taking into account information from the linewidth of the signal and the metabolite basis set.
Download

Paper Nr: 57
Title:

UNSUPERVISED 3D SEGMENTATION OF HIPPOCAMPUS IN BRAIN MR IMAGES

Authors:

Sandeep S. Kaushik and Jayanthi Sivaswamy

Abstract: The most widely followed procedure for diagnosis and prognosis of dementia is structural neuroimaging of hippocampus by means of MR. Hippocampus segmentation is of wide interest as it enables quantitative assessment of the structure. In this paper, we propose an algorithm for hippocampus segmentation that is unsupervised and image driven. It is based on a hybrid approach which combines a coarse segmentation and surface evolution. A coarse solution is derived using region growing which is further refined using a modified version of the physics based water flow model (Liu and Nixon, 2007). The proposed method has been tested on a publicly available dataset. The performance of this method is assessed using Dice coefficient against the ground truth provided for 25 volume images. It is consistent across volumes and the average Dice values are comparable to a multi-atlas based method reported on a subset of the same dataset.
Download

Paper Nr: 61
Title:

USING GRANGER CAUSALITY TO CHARACTERISE BIDIRECTIONAL INTERACTIONS IN THE HUMAN BRAIN DURING INDUCTION OF ANAESTHESIA

Authors:

Nicoletta Nicolaou, Julius Georgiou, Saverios Houris and Pandelitsa Alexandrou

Abstract: General anaesthesia is a reversible state whereby conscious experience is disrupted and reflexes to afferent stimuli are depressed. The precise method of action of anaesthetic agents is still largely unknown. However, the administration of anaesthetics causes observable changes in the electrical brain activity (EEG), the study of which can provide an insight into the mechanism of action of general anaesthesia. This paper investigates the patterns of bidirectional interactions that are manifest in brain activity during anaesthetic induction with propofol. Granger Causality is applied to the EEG of patients scheduled for surgery under general anaesthesia as a means of characterising the interactions between different brain areas prior and after the administration of the anaesthetic agents. Strong unidirectional information flow between frontal and posterior areas was found to occur shortly after anaesthetic induction.
Download

Paper Nr: 62
Title:

HEMODYNAMIC FEATURES EXTRACTION FROM A NEW ARTERIAL PRESSURE WAVEFORM PROBE

Authors:

V. G. Almeida, P. Santos, E. Figueiras, E. Borges, T. Pereira, J. Cardoso, C. Correia and H. C. Pereira

Abstract: In this work, we discuss an algorithm that reliable and accurately identifies the prominent points of the cardiac cycle: the systolic peak (SP), reflection point (RP), dicrotic notch, (DN) and dicrotic peak (DP). The prominent point’s identifier algorithm (PPIA) action is based on the analysis a number of features of the arterial pressure waveform and its first derivative, and is part of the fundamental software analysis pack for a new piezoelectric probe designed to reproduce the arterial pressure waveform from the pulsatile activity taken non-invasively at the vicinity of a superficial artery. The output PPIA is the coordinates (in time and amplitude) of the above referred points. To assess the accuracy of the algorithm, a reference database of 173 pulses from eight volunteers, was established and the values yielded by the PPIA were compared to annotations from a human expert engineer (HEE). The quality of the results is statistically quantified either in time as in amplitude. Average values of 4.20% for error, 99.09% for sensitivity and 96.77% for positive predictive value were found to be associated to time information while amplitude yields averages of 2.68%, 99.08% and 98.22%, respectively, for the same parameters.
Download

Paper Nr: 65
Title:

CLASSIFICATION OF HUMAN PHYSICAL ACTIVITIES FROM ON-BODY ACCELEROMETERS - A Markov Modeling Approach

Authors:

Andrea Mannini and Angelo Maria Sabatini

Abstract: Several applications demanding the development of small networks of on-body sensors, such as motion sensors, are currently investigated. Accelerometers are a popular choice as motion sensors: the reason is partly in their capability of extracting information that can be used to automatically infer the physical activity the human subject is involved, beside their role in feeding estimators of biomechanical parameters. Automatic classification of human physical activities is highly attractive for pervasive computing systems, whereas contextual awareness may ease the human-machine interaction, and in biomedicine, whereas wearable sensor systems are proposed for long-term monitoring of physiological and biomechanical parameters. This paper is concerned with the machine learning algorithms needed to perform the classification task. Hidden Markov Model (HMM) classifiers are studied by contrasting them with Gaussian Mixture Model (GMM) classifiers. HMMs incorporate the statistical information available on movement dynamics into the classification process, without discarding the time history of previous outcomes, as GMMs do. In this work, rather than considering them as models for single motor activities, we apply HMMs as models suitable for sequences of chained activities. An example of the benefits of the statistical leverage by HMMs is illustrated and discussed by analyzing a dataset of accelerometer time series.
Download

Paper Nr: 66
Title:

A FRACTAL-BASED ALGORITHM OF EMOTION RECOGNITION FROM EEG USING AROUSAL-VALENCE MODEL

Authors:

Olga Sourina and Yisi Liu

Abstract: Emotion recognition from EEG could be used in many applications as it allows us to know the “inner” emotion regardless of the human facial expression, behaviour, or verbal communication. In this paper, we proposed and described a novel fractal dimension (FD) based emotion recognition algorithm using an Arousal-Valence emotion model. FD values calculated from the EEG signal recorded from the corresponding brain lobes are mapped to the 2D emotion model. The proposed algorithm allows us to recognize emotions that could be defined by arousal and valence levels. Only 3 electrodes are needed for the emotions recognition. Higuchi and box-counting algorithms were used for the EEG analysis and comparison. Support Vector Machine classifier was applied for arousal and valence levels recognition. The proposed method is a subject dependent one. Experiments with music and sound stimuli to induce human emotions were realized. Sound clips from the International Affective Digitized Sounds (IADS) database were used in the experiments.
Download

Paper Nr: 71
Title:

AUTOMATIC SEGMENTATION OF CONDUCTIVITY CHANGES IN ELECTRICAL IMPEDANCE TOMOGRAPHY IMAGES

Authors:

A. Zifan, P. Liatsis, P. Kantartzis and R. Vargas-Canas

Abstract: In this paper, we propose a novel method for the automatic segmentation of Electrical Impedance Tomography (EIT) lung images. EIT is a non-invasive technique, which produces low-spatial and high-temporal resolution images of the internal resistivity of the region of the body probed by currents. EIT is the only technology that reliably quantifies regional lung volumes non-invasively. The problem is non-linear and ill-conditioned and can be solved using 2D or 3D finite element methods (FEMs) subject to using appropriate regularisation strategies. The usual method of segmenting EIT lung images is to manually select a region of interest and derive statistical measures. This procedure is not suitable for FEM-based models as it works on rectangular pixels, as well as making the task tedious and time consuming. We propose an alternative segmentation framework, which operates directly on the resulting FEM meshes, prior to rasterisation in order to prevent the propagation of errors in the reconstructed resistivity regions, due to mapping onto a rectangular grid. We use a spatio-temporal probabilistic method to segment conductivity changes in the EIT thorax images. Application of the proposed method offers a much needed alternative to interactive segmentation currently favoured by EIT researchers and clinicians.
Download

Paper Nr: 80
Title:

TOWARDS A STATISTICAL DESCRIPTION OF EXPERIMENTAL DATA FOR DETECTION-ESTIMATION PROBLEMS IN DNA TRANSLOCATIONS THROUGH NANOPORES

Authors:

S. Michelet, J-P. Barbot, O. Francais, P-Y. Joubert, P. Larzabal, R. Kawano, H. Sasaki, T. Osaki, S. Takeuchi and B. Le Pioufle

Abstract: This paper investigates the properties of DNA translocations signals in a stochastic framework. The considered signals are relative to the translocation of single strand DNA through natural nanopores, and are obtained using a planar patch clamp method. The stochastic signal analysis is carried out considering the statistical distribution of DNA translocation parameters, considered as random variables including the amplitude, the duration and the apparition of the DNA translocation events as well as the no-translocation signal features. For each of these variables, a distribution function is proposed and assessed using a Kolmogorov-Smirnov test, and their features are estimated. The DNA translocation signal stochastic analysis enables to characterize the detection and/or estimation performances of existing algorithms, such as a breakdown detection algorithm, in a stochastic framework. Moreover, it opens the way to the design of model based algorithms such as detection tests using a likelihood ratio or joint detection-estimation algorithms using a maximum likelihood approach, for an enhanced characterization of DNA translocations.
Download

Paper Nr: 82
Title:

A NEW P300 NO EYE-GAZE BASED INTERFACE: GEOSPELL

Authors:

F. Aloise, P. Aricò, F. Schettini, A. Riccio, M. Risetti, S. Salinari, D. Mattia, F. Babiloni and F. Cincotti

Abstract: Brain Computer Interface (BCI) is an alternative communication system which allows users to send commands and/or messages toward the outside not crossing the normal output channels of the brain, but conveying these outputs from the human brain to a computer (Wolpaw et al., 2002). In an EEG-based BCI messages are obtained from brain activity. This study presents a novel P300 based Brain Computer Interface requiring no eye gaze, and so usable in covert attention status, called GeoSpell (Geometric Speller). GeoSpell performances have been compared with those obtained by the subjects with the standard 6 by 6 P300 Speller (P3Speller) matrix which depends on eye gaze (Farwell and Donchin, 1988). A NASA Task Load Index (TLX) workload assessment was employed to provide a subjective rating about the task’s workload and satisfaction with respect to both the interfaces (NASA Human Performance Research Group 1987). Results shown comparable workload values for P3Speller and Geospell; this result has an important impact in term of efficiency and satisfaction for the use of the BCI devices. Geospell interface has shown an accuracy comparable with the P3Speller one but with a lower bit-rate.
Download

Paper Nr: 84
Title:

RETINAL VASCULAR NETWORK MODEL - An Automatic Approach

Authors:

Alauddin Bhuiyan, Baikunth Nath and Rao Kotagiri

Abstract: In this paper, we propose a retinal vascular network model, which is an automatic process of generating a graph representation (i.e., a tree) of the retinal blood vessels and includes vessel geometrical features. It maps the retinal blood vessels and can facilitate vascular features such as the vessel width, bifurcation angle, among others to predict or earlier diagnose cardiovascular and related diseases. The proposed tree-model is based on vessel’s centerline, cross-sectional width, and bifurcation, branching and crossover points. The optic disc center is computed using the Hough transformation and vessel centerlines are tracked from out side its radius. Blood vessels are fragmented as vessel-segments based on the bifurcation, branching and crossover points. For each blood vessel we construct a binary tree which is linked in the root of the tree-model. Our automated method achieves an accuracy of 91.23% in extracting the vessel-segments.
Download

Paper Nr: 85
Title:

TOWARD DOMOTIC APPLIANCES CONTROL THROUGH A SELF-PACED P300-BASED BCI

Authors:

F. Aloise, F. Schettini, P. Aricò, F. Leotta, S. Salinari, D. Mattia, F. Babiloni and F. Cincotti

Abstract: During recent years there has been a growing interest in Brain Computer Interface (BCI) systems as an alternative means of interaction with the external world for people with severe motor disabilities. The use of the P300 event-related potentials as control feature allows users to choose between various options (letters or icons) requiring a very short calibration phase. The aim of this work is to improve performances and flexibility of P300 based BCIs. An efficient BCI system should be able to understand user's intentions from the ongoing EEG, abstaining from doing a selection when the user is engaged in a different activity, and changing its speed of selection depending on current user's attention level. Our self-paced system addresses all these issues representing an important step beyond the classical synchronous P300 BCI that forces the user in a continuous control task. Experimentation has been performed on 10 healthy volunteers acting on a BCI-controlled domestic environment in order to demonstrate the potential usability of BCI systems in everyday life. Results show that the self-paced BCI increases information transfer rate with respect to the synchronous one, being very robust, at the same time, in avoiding false negatives when the user is not engaged in a control task.
Download

Paper Nr: 87
Title:

DEVELOPMENT OF WEARABLE GAIT EVALUATION SYSTEM - A Preliminary Test of Measurement of Joint Angles and Stride Length

Authors:

Takashi Watanabe, Hiroki Saito, Eri Koike and Kazuki Nitta

Abstract: The purpose of this study is to develop wearable sensor system for gait evaluation using gyroscopes and accelerometers for application to rehabilitation, healthcare and so on. In this paper, simultaneous measurement of joint angles of the lower limbs and stride length was tested with a prototype of wearable sensor system. The system measured the joint angles using the Kalman filter. Signals from the sensor attached on the foot were used in the stride length estimation detecting foot movement automatically. Joint angles of the lower limbs and the stride length were measured with reasonable accuracy compared to those values measured with optical motion measurement system with healthy subjects. Joint angle patterns measured in 10m walking with a healthy subject were similar to common patterns. High correlation between joint angles at some characteristic points and walking speed were also found adequately from measured data. The system was suggested to be able to detect characteristics of gait.
Download

Paper Nr: 93
Title:

OPTIMIZING THE ASSESSMENT OF CEREBRAL AUTOREGULATION FROM LINEAR AND NONLINEAR MODELS

Authors:

Natalia Angarita Jaimes and David Simpson

Abstract: Autoregulation mechanisms maintain blood flow approximately stable despite changes in arterial blood pressure. Mathematical models that characterize this system have been used in the quantitative assessment of function/impairment of autoregulation as well as in furthering the understanding of cerebral hemodynamics. Using spontaneous fluctuations in arterial blood pressure (ABP) as input and cerebral blood flow velocity (CBFV) as output, the autoregulatory mechanism has been modeled using linear and nonlinear approaches. From these models, a small number of measures have been extracted to provide an overall assessment of autoregulation. Previous studies have considered a single – or at most- a couple of measures, making it difficult to compare the performance of different autoregulatory parameters (and the different modeling approaches) under similar conditions. We therefore compare the performance of established autoregulatory parameters in addition to novel features extracted from the models’ response to a band-pass filtered impulse. We investigate if some of the poor performance previously reported can be overcome by a better choice of autoregulation parameter to extract from the model. Twenty-six recordings of ABP and CBFV from normocapnia and hypercapnia in 13 healthy adults were analyzed. In the absence of a ‘gold’ standard for the study of dynamic cerebral autoregulation, lower inter and intra subject variability of the parameters and better separation between normo- and hyper-capnia states were considered as criteria for identifying improved measures of autoregulation. We found that inter- and intra- subject variability in the assessment of autoregulation can be significantly improved by a careful choice of autoregulation measure extracted from either linear or non-linear models.
Download

Paper Nr: 94
Title:

TWO-MODES CYCLIC BIOSIGNAL CLUSTERING BASED ON TIME SERIES ANALYSIS

Authors:

Neuza Nunes, Tiago Araújo and Hugo Gamboa

Abstract: In this paper we introduce an unsupervised learning algorithm which distinguishes two different modes in a cyclic signal. We also present the concept of “mean wave” which averages all signal waves aligned in a notable point (nth zero derivative). With that information the signal’s morphology is captured. The clustering mechanism is based on the information collected with the mean wave approach using a k-means algorithm. The algorithm produced is signal-independent, and therefore can be applied to any type of signal providing it is a cyclic signal that has no major changes in the fundamental frequency. To test the effectiveness of the proposed method, we acquired several biosignals (accelerometry, electromyography and blood volume pressure signals) in the context tasks performed by the subjects with two distinct modes in each. The algorithm successfully separates the two modes with 99.2% of efficiency. The fact that this approach doesn’t require any prior information and the preliminary good classification performance makes this algorithm a powerful tool for biosignals analysis and classification.
Download

Paper Nr: 99
Title:

INTELLIGIBILITY OF ELECTROLARYNX SPEECH USING A NOVEL HANDS-FREE ACTUATOR

Authors:

Brian Madden, Mark Nolan, Edward Burke, James Condron and Eugene Coyle

Abstract: During voiced speech, the larynx provides quasi-periodic acoustic excitation of the vocal tract. In most electrolarynxes, mechanical vibrations are produced by a linear electromechanical actuator, the armature of which percusses against a metal or plastic plate at a frequency within the range of glottal excitation. In this paper, the intelligibility of speech produced using a novel hands-free actuator is compared to speech produced using a conventional electrolarynx. Two able-bodied speakers (one male, one female) performed a closed response test containing 28 monosyllabic words, once using a conventional electrolarynx and a second time using the novel design. The resulting audio recordings were randomized and replayed to ten listeners who recorded each word that they heard. The results show that the speech produced using the hands-free actuator was substantially more intelligible to the majority of listeners than that produced using the conventional electrolarynx. The new actuator has properties (size, weight, shape, cost) which lends itself as a suitable candidate for possible hands-free operation. This is one of the research ideals for the group and this test methodology presented as a means of testing intelligibility. This paper outlines the procedure for the possible testing of intelligibility of electrolarynx designs.
Download

Paper Nr: 100
Title:

THE DYNAMICS OF LOCUST NON-SPIKING LOCAL INTERNEURONS - Responses to Imposed Limb Movements

Authors:

Oliver P. Dewhirst, Natalia Angarita-Jaimes, David M. Simpson, Robert Allen and Philip L. Newland

Abstract: A key feature of the locusts hind leg control system is a reflex loop that uses a stretch sensor, the Femoral Chordotonal organ, to monitor the position and movements of the tibia relative to the femur. A population of non-spiking local interneurons in the metathoracic ganglia receive synaptic inputs from the sensory neurons in the chordotonal organ and indirect inputs from other interneurons. They function to integrate these signals and generate the motor pattern required for coordinated limb movement. Nonlinear Volterra models combined with Gaussian white noise stimulation have, for the first time, been used to characterise the dynamics of this population of interneurons. The results show that the interneurons can be clustered into three groups, those which are position, position/velocity and velocity sensitive.
Download

Paper Nr: 101
Title:

WBAN BASED PROTOTYPE FOR ACTIVE BODY CLIMATE CONTROL BASED ON ENVIRONMENTAL AND INDIVIDUAL SENSOR DATA

Authors:

A. Gharbi, M. Breuel, W. Darmoul, W. Stork, K. D. Mueller-Glaser, S. Heuer and S. Haertel

Abstract: In this paper, a new textile integrated WBAN based prototype for active body climate control is presented. The design of this prototype resulted from a previous evaluation system, which has been tested in a field study. In addition, new algorithms for determining the metabolic activity from measured vital parameters (heart rate and activity) and thus controlling the cooling mechanism by setting the necessary ventilation airflow have been conceived. A field study involving nine test persons has been conducted in order to test the new prototype and validate the conceived algorithms.
Download

Paper Nr: 102
Title:

SPECTRAL ANALYSIS OF THE CEREBRAL ACTIVITY DURING VOLUNTARY MODULATION OF MENTAL STATES - A High Resolution EEG Study

Authors:

J. Toppi, F. Babiloni, F. Cincotti, F. De Vico Fallani, G. Vecchiato, S. Salinari, D. Mattia and L. Astolfi

Abstract: In the neuroscience field, the use of advanced techniques of EEG recording and analysis, led to look for adequate methodology to prevent type I errors, which occur in computing thousands of univariate tests in order to highlight the brain areas in which significant activity arises. In this paper we illustrate the capability of tracking the brain activity during tasks consisting in tennis playing imagery and spatial navigation imagery, by using advanced high resolution EEG methodology accompanied by the use of appropriate statistical techniques that takes into account the risk of the Type I errors. Results showed that in the Spatial Navigation condition the power spectra activity is significantly different from the rest in the bilateral parietal areas and left motor area, while in the Tennis condition the cortical activity differs from the rest in bilateral parietal areas and in the left sensory-motor cortex. These preliminary findings are in partial accordance with previous hemodynamic studies.
Download

Paper Nr: 104
Title:

ECG P-WAVE SMOOTHING AND DENOISING BY QUADRATIC VARIATION REDUCTION

Authors:

Antonio Fasano, Valeria Villani, Luca Vollero and Federica Censi

Abstract: Atrial fibrillation is the most common persistent cardiac arrhythmia and it is characterized by a disorganized atrial electrical activity. Its occurrence can be detected, and even predicted, through P-waves time-domain and morphological analysis in ECG tracings. Given the low signal-to-noise ratio associated to P-waves, such analysis are possible if noise and artifacts are effectively filtered out from P-waves. In this paper a novel smoothing and denoising algorithm for P-waves is proposed. The algorithm is solution to a convex optimization problem. Smoothing and denoising are achieved reducing the quadratic variation of the measured P-waves. Simulation results confirm the effectiveness of the approach and show that the proposed algorithm is remarkably good at smoothing and denoising P-waves. The achieved SNR gain exceeds 15 dB for input SNR below 6 dB. Moreover the proposed algorithm has a computational complexity that is linear in the size of the vector to be processed. This property makes it suitable also for real-time applications.
Download

Paper Nr: 105
Title:

SESSION-INDEPENDENT EMG-BASED SPEECH RECOGNITION

Authors:

Michael Wand and Tanja Schultz

Abstract: This paper reports on our recent research in speech recognition by surface electromyography (EMG), which is the technology of recording the electric activation potentials of the human articulatory muscles by surface electrodes in order to recognize speech. This method can be used to create Silent Speech Interfaces, since the EMG signal is available even when no audible signal is transmitted or captured. Several past studies have shown that EMG signals may vary greatly between different recording sessions, even of one and the same speaker. This paper shows that session-independent training methods may be used to obtain robust EMG-based speech recognizers which cope well with unseen recording sessions as well as with speaking mode variations. Our best session-independent recognition system, trained on 280 utterances of 7 different sessions, achieves an average 21.93% Word Error Rate (WER) on a testing vocabulary of 108 words. The overall best session-adaptive recognition system, based on a session-independent system and adapted towards the test session with 40 adaptation sentences, achieves an average WER of 15.66%, which is a relative improvement of 21% compared to the baseline average WER of 19.96% of a session-dependent recognition system trained only on a single session of 40 sentences.
Download

Paper Nr: 109
Title:

PREREQUISITES FOR AFFECTIVE SIGNAL PROCESSING (ASP) - Part V - A Response to Comments and Suggestions

Authors:

Egon L. van den Broek, Joris H. Janssen, Marjolein D. van der Zwaag, Joyce H. D. M. Westerink and Jennifer A. Healey

Abstract: In four papers, a set of eleven prerequisites for affective signal processing (ASP) were identified (van den Broek et al., 2010): validation, triangulation, a physiology-driven approach, contributions of the signal processing community, identification of users, theoretical specification, integration of biosignals, physical characteristics, historical perspective, temporal construction, and real-world baselines. Additionally, a review (in two parts) of affective computing was provided. Initiated by the reactions on these four papers, we now present: i) an extension of the review, ii) a post-hoc analysis based on the eleven prerequisites of Picard et al.(2001), and iii) a more detailed discussion and illustrations of temporal aspects with ASP.
Download

Paper Nr: 111
Title:

OPTICAL FLOW BY MULTI-SCALE ANNOTATED KEYPOINTS - A Biological Approach

Authors:

Miguel Farrajota, J. M. F. Rodrigues and J. M. H. du Buf

Abstract: Optical flow is the pattern of apparent motion of objects in a visual scene and the relative motion, or egomotion, of the observer in the scene. In this paper we present a new cortical model for optical flow. This model is based on simple, complex and end-stopped cells. Responses of end-stopped cells serve to detect keypoints and those of simple cells are used to detect orientations of underlying structures and to classify the junction type. By combining a hierarchical, multi-scale tree structure and saliency maps, moving objects can be segregated, their movement can be obtained, and they can be tracked over time. We also show that optical flow at coarse scales suffices to determine egomotion. The model is discussed in the context of an integrated cortical architecture which includes disparity in stereo vision.
Download

Paper Nr: 116
Title:

ON SPEECH RECOGNITION PERFORMANCE UNDER NON-STATIONARY ECHO CANCELLATION

Authors:

Mahdi Triki

Abstract: During the last decades, performance of speech recognizers significantly increased for large vocabulary tasks and adverse environments. To reduce interference, acoustic echo cancellation has been proposed and extensively investigated. Particular attention was paid to the convergence proprieties and the capability to handle double talk. However, in time-varying environment, the echo canceller has the additional task to track the variations of the propagation channel. With this respect, it has been established that algorithms that exhibit fast convergence do not provide necessarily good tracking performances. In such an environment, performance assessment is also challenging and the ‘experiment’ design is crucial to provide consistent and interpretable results. In the present paper, we reproduce time-varying artifacts by altering the surrounding acoustic environment (using a moving person/robot). The movement characteristics (discrete/continuous) and location (line-of-sight/background) emphasizes different room/algorithms characteristics and provides deeper insights on the system behavior.
Download

Paper Nr: 117
Title:

STATISTICAL ANALYSIS OF THE SIGNAL AND PROSODIC SIGN OF COGNITIVE IMPAIRMENT IN ELDERLY-SPEECH - A Preliminary Study

Authors:

Shohei Kato, Yuta Suzuki, Akiko Kobayashi, Toshiaki Kojima, Hidenori Itoh and Akira Homma

Abstract: This paper presents a novel approach for early detection of cognitive impairment in the elderly. Our approach incorporates the use of speech sound analysis and multivariate statistical techniques. In this paper, we focus on the prosodic features of speech. Fifty six Japanese subjects (22 males and 34 females between the ages of 64 and 90 years) participated in this study. Blind to clinical groups, we collected speech sounds from segments of dialogue during an HDS-R examination. The segments corresponds to speech sounds from answers to questions about time orientation and number backward counting. Ninety eight prosodic features were extracted from each of the speech sounds. These prosodic features consisted of spectral and pitch features (13), formant features (61), intensity features (22), and speech rate and response time (2). These features were refined by principal component analysis and/or feature selection. In addition, we calculated speech prosody-based cognitive impairment rating (SPCIR) by multiple linear regression analysis. The results indicate that a moderately significant correlation exists between the HDS-R score and the synthesis of several selected prosodic features. Consequently, the adjusted coefficient of determination (R2 = 0.57) suggests that prosody-based speech sound analysis could potentially be used to detect cognitive impairment in elderly subjects.
Download

Paper Nr: 125
Title:

COMPARISON OF LINEAR CLASSIFICATION METHODS FOR P300 BRAIN-COMPUTER INTERFACE ON DISABLED SUBJECTS

Authors:

Nikolay V. Manyakov, Nikolay Chumerin, Adrien Combaz and Marc M. Van Hulle

Abstract: In this paper, we investigate the accuracy of linear classification techniques for a P300 Brain-Computer Interface used in a typing paradigm. Fisher’s Linear Discriminant Analysis (LDA), Bayesian Linear Discriminant Analysis (BLDA), Stepwise Linear Discriminant Analysis (SLDA), linear Support Vector Machine (SVM) and a method based on Feature Extraction (FE) were compared. Experiments were performed on patients suffering from Amyotrophic Lateral Sclerosis (ALS), middle cerebral artery (MCA) stroke and Subarachnoid Hemorrhage (SAH), in on-line and off-line mode. Our results show that BLDA yields a significantly higher accuracy than the other linear techniques we have compared, at least for our group of subjects.
Download

Paper Nr: 128
Title:

GAZE TRAJECTORY AS A BIOMETRIC MODALITY

Authors:

Farzin Deravi and Shivanand Guness

Abstract: Could everybody be looking at the world in a different way? This paper explores the idea that every individual has a distinctive way of looking at the world and thus it may be possible to identify an individual by how they look at external stimuli. The paper reports on a project to assess the potential for a new biometric modality based on gaze. A gaze tracking system was used to collect gaze information of participants while viewing a series of images for about 5 milliseconds each. The data collected was firstly analysed to select the best suited features using three different algorithms: the Forward Feature Selection, the Backwards Feature Selection and the Branch and Bound Feature Selection algorithms. The performance of the proposed system was then tested with different amounts of data used for classifier training. From the preliminary experimental results obtained, it can be seen that gaze does have some potential as being used as a biometric modality. The experiments carried out were only done on a very small sample; more testing is required to confirm the preliminary findings of this paper.
Download

Paper Nr: 129
Title:

AN AUTOMATED METHOD FOR RETINAL IMAGE MATCHING USING VASCULAR FEATURES

Authors:

Alauddin Bhuiyan, Baikunth Nath, Kotagiri Ramamohanarao and Tien Y. Wong

Abstract: In this paper, we propose a method for retinal image matching that can be used in image matching for person identification or patient longitudinal study. Vascular invariant features are extracted from the retinal image and a feature vector is constructed for each of the vessel-segments in the retinal blood vessels. The feature vectors are represented in a tree structure with maintaining the vessel-segments actual hierarchical positions. Using these feature vectors, corresponding images are matched. An image matching method is demonstrated which identifies the same vessel in the corresponding images for comparing the desired feature(s). Initial results demonstrate that the method is suitable for image matching and patient longitudinal study.
Download

Paper Nr: 146
Title:

TOWARDS A FINGER BASED ECG BIOMETRIC SYSTEM

Authors:

André Lourenço, Hugo Silva, Daniel Perna Santos and Ana Fred

Abstract: The ECG signal has been shown to contain relevant information for human identification. Even though results validate the potential of these signals, data acquisition methods and apparatus explored so far compromise user acceptability. In this paper we propose an ECG based biometric system that uses signals collected at the fingers through a minimally intrusive 1-lead ECG setup. Time domain ECG signal processing is performed, following the usual steps of filtering, peak detection, heartbeat waveform segmentation, and amplitude normalization. We introduce two additional steps of synthetic waves generation and time normalization. Through a simple one nearest neighbor classifier, results have revealed this to be a promising technique.
Download

Paper Nr: 153
Title:

ECG-BASED CONTINUOUS AUTHENTICATION SYSTEM USING ADAPTIVE STRING MATCHING

Authors:

David Pereira Coutinho, Ana L. N. Fred and Mario A. T. Figueiredo

Abstract: Conventional access control systems are typically based on a single time instant authentication. However, for high-security environments, continuous user verification is needed in order to robustly prevent fraudulent or unauthorized access. The electrocardiogram (ECG) is an emerging biometric modality with the following characteristics: (i) it does not require liveliness verification, (ii) there is strong evidence that it contains sufficient discriminative information to allow the identification of individuals from a large population, (iii) it allows continuous user verification. Recently, a string matching approach for ECG-based biometrics, using the Ziv-Merhav (ZM) cross parsing, was proposed. Building on previous work, and exploiting tools from data compression, this paper goes one step further, proposing a method for ECG-based continuous authentication. An adaptive way of using the ZM cross parsing is introduced. The use of the Lloyd-Max quantization is also introduced to improve the results with the string matching approach for ECG-based biometrics. Results on one-lead ECG real data are presented, acquired during a concentration task, from 19 healthy individuals.
Download

Paper Nr: 154
Title:

IMPROVING LEARNING ABILITY OF RECURRENT NEURAL NETWORKS - Experiments on Speech Signals of Patients with Laryngopathies

Authors:

Jarosław Szkoła, Krzysztof Pancerz and Jan Warchoł

Abstract: Recurrent neural networks can be used for pattern recognition in time series data due to their ability of memorizing some information from the past. The Elman networks are a classical representative of this kind of neural networks. In the paper, we show how to improve learning ability of the Elman network by modifying and combining it with another kind of a recurrent neural network, namely, with the Jordan network. The modified Elman-Jordan network manifests a faster and more exact achievement of the target pattern. Validation experiments were carried out on speech signals of patients with laryngopathies.
Download

Paper Nr: 155
Title:

NEONATAL SEIZURE DETECTION USING BLIND ADAPTIVE FUSION

Authors:

Huaying Li and Aleksandar Jeremic

Abstract: Seizure is the result of excessive electrical discharges of neurons, which usually develops synchronously and happens suddenly in the central nervous system. Clinically, it is difficult for physician to identify neonatal seizures visually, while EEG seizures can be recognized by the trained experts. Usually, in NICUs, EEG monitoring systems are used in stead of the expensive on-site supervision. However, it is a waste of time to review an overnight recording, which motivates the researchers to develop automated seizure detection algorithms. Although, there are few detection algorithms existed in the literature, it is difficult to evaluate these mathematical model based algorithm since their performances vary significantly on different data sets. By extending our previous results on multichannel information fusion, we propose a distributed detection system consisting of the existing detectors and a fusion center to detect the seizure activities in the newborn EEG. The advantage of this proposed technique is that it does not require any priori knowledge of the hypotheses and the detector performances, which are often unknown in real applications. Therefore, this proposed technique has the potential to improve the performances of the existing neonatal seizure detectors. In this paper, we first review two newborn EEGmodels, one of which is used to generate neonatal EEG signals. The synthetic data is used later for testing purpose. We also review three existing algorithms on this topic and implement them to work as the local detectors of the system. Then, we introduce the fusion algorithms applied in the fusion center for two different scenarios: large sample size and small sample size. We finally provide some numerical results to show the applicability, effectiveness, and the adaptability of the blind algorithms in the seizure detection problem. We also provide the testing results obtained using the synthetic to show the improvement of the detection system.
Download

Paper Nr: 160
Title:

A ROBUST AND PRACTICAL METHOD TO SEPARATE PERIODIC SIGNALS FROM MEG DATA USING SECOND ORDER STATISTICS

Authors:

Hidekazu Fukai

Abstract: The analyses of recordings of magnetoencephalography (MEG) and other imaging techniques may require the separation of periodic signals from the observed signals. Blind source separation (BSS) is widely used for the separation of specific signals these days. Though several algorithms based on the BSS scheme for the separation of periodic signals have been proposed, they usually assume the system to be well-posed, satisfactory results often cannot be obtained for practical recordings. In this study, we show that a method based on the joint approximate diagonalization of correlation matrices with several time delays (JADCM) is robust and good results can be obtained by choosing the time delays carefully, especially in practical ill-posed situations such as signal separation from MEG recordings. The performance of the proposed method is compared with that of Periodic BSS and JADCM using the conventional parameter set.
Download

Paper Nr: 165
Title:

MULTITASK LEARNING APPLIED TO SPATIAL FILTERING IN MOTOR IMAGERY BCI - A Preliminary Offline Study

Authors:

Dieter Devlaminck, Bart Wyns, Georges Otte and Patrick Santens

Abstract: Motor imagery based brain-computer interfaces (BCI) commonly use the common spatial pattern filter (CSP) as preprocessing step before feature extraction and classifiction. The CSP method is a supervised algorithm and therefore needs subject specific training data for calibration, which is very time consuming to collect. Instead of letting all that data and effort go to waste, the data of other subjects could be used to further improve results for new subjects. This problem setting is often encountered in multitask learning, from which we will borrow some ideas and apply it to the preprocessing phase. This paper outlines the details of the multitask CSP algorithm and shows some results on data from the third BCI competition. In some of the subjects a clear improvement can be seen by using information of other subjects, while in some subjects the algorithm determines that a specific model is the best. We also compare the use of a global filter, which is constructed only with data of other subjects, with the case where we ommit any form of spatial filtering. Here, the global filter seems to boost performance in four of the five subjects.
Download

Paper Nr: 8
Title:

WHICH RESOLUTION FOR RELIABLE ECG P-WAVE ANALYSIS IN ATRIAL FIBRILLATION?

Authors:

Federica Censi, Giovanni Calcagnini, Michele Triventi, Eugenio Mattei, Pietro Bartolini, Ivan Corazza and Giuseppe Boriani

Abstract: P-wave analysis is becoming more and more used to help indentifying patients at risk for AF. Particularly, precise measurement of P-wave duration is an important factor in determining the risk of atrial arrhythmias. However, the methods to extract P-wave duration must be precise and reliable. Automatic analysis of P-wave must take into account technical aspects, one of those being the bit resolution. The aim of this manuscript is to investigate the effects of amplitude resolution of ECG acquisition systems on P-wave analysis. Starting from ECG recorded by an acquisition system with a LSB of 31 nV (24-bit on an input range of 524mVpp), we reproduced ECG signal as acquired by systems with lower resolution (16, 15, 14, 13 and 12 bit). We found that, when LSB is of the order of 128 µV (12 bit), a single P-wave is not recognizable on ECG (figure 1, upper panel). However, when averaging is applied, a P-wave template can be extracted, apparently suitable for P-wave analysis. Results obtained in terms of P-wave duration revealed that at lowest resolution (from 12 to 14 bit) the error on P-wave duration estimation is important and could lead to misleading results. However, the resolution used nowadays in modern electrocardiographs (15 and 16 bit) lead to results rather similar to those obtained with higher resolution.
Download

Paper Nr: 12
Title:

NON UNIFORM GEOMETRY BEND SENSORS EXPLOITED FOR BIOMEDICAL SYSTEMS

Authors:

Giovanni Saggio, Stefano Bocchetti, Carlo Alberto Pinto, Giuseppe Latessa and Giancarlo Orengo

Abstract: In biomedical systems the bend sensors have been increasingly used stands their interesting properties useful to measure human joint static and dynamic postures. These commercially available sensors are usually made of a polyester film printed on with a special carbon ink. The film acts as a support while the ink’s resistance value changes with bending dues to an applied external force. The substrate film material is usually made by Kapton and/or Mylar for their properties, stands the fact that substrate must be able to bend repeatedly without failure for the sensor to work. In spite of their interesting properties the commercial bend sensors have a resistance vs. bent angle characteristic which is not actually ideal as a linear function, to measure human postures, would be. So we introduce here a novel solution useful to linearize the sensor response.
Download

Paper Nr: 14
Title:

INVESTIGATION OF CHANGES IN KINETIC TREMOR THROUGH ANALYSIS OF HAND-DRAWING MOVEMENTS - Differences between Physiological and Essential Tremors

Authors:

Maria Fernanda S. Almeida, Guilherme L. Cavalheiro, Adriano O. Andrade, Daniel A. Furtado and Adriano A. Pereira

Abstract: Tremor is the most common movement disorder characterized by repetitive and stereotyped movements. Tremor can be classified in many ways, depending on its phenomenology, frequency and location. The data collection conducted under kinetic conditions and while performing a voluntary movement highlights the kinetic tremor. The analysis of hand-drawing movements is commonly used in the evaluation of patients with tremor. In this study, a number of features extracted from tremor activity, obtained from digitized drawings of Archimedes’ spirals, were analysed. The analyses followed the sequence bellow: 1 – Linearization of the spiral of Archimedes; 2 – Estimate of tremor activity; 3 – Data pre-processing; and 4 – Feature extraction from the tremor activity. The statistical analysis of the extracted features was able to prove the differences between physiological and essential tremors collected under kinetic conditions.
Download

Paper Nr: 26
Title:

BIOSONAR-INSPIRED SOURCE LOCALIZATION IN LOW SNR

Authors:

Sasha Apartsin, Leon N. Cooper and Nathan Intrator

Abstract: Some mammals use sound signals for communications and navigation in the air (bats) or underwater (dolphins). Recent biological discovery shows that blind mole rat is capable of detecting and avoiding underground obstacles using reflection from seismic signals. Such a remarkable capacity relies on the ability to localize the source of the reflection with high accuracy and in very low Signal to Noise Ratio (SNR) conditions. The standard methods for source localization are usually based on Time of Arrival (ToA) estimation obtained by the correlation of received signal with a matched filter. This approach suffers from rapid deterioration in the accuracy as SNR level falls below certain threshold value: the phenomenon known in the Radar Theory as a “threshold effect”. In this paper we describe biosonar-inspired method for ToA estimation and 2D source localization based on the fusion of the measurements from biased estimators which are obtained using a family of unmatched filters. Suboptimal but not perfectly correlated estimators are combined together to produce a robust estimator for ToA and 2D source position which outperforms standard matched filter-based estimator in high noise. The proposed method can be applied for mapping of underground instalments using low power infrasound pulses.
Download

Paper Nr: 34
Title:

DIASTOLIC TIMED VIBRATIONS FOR PRE-HOSPITALIZATION TREATMENT OF MYOCARDIAL INFARCTION

Authors:

Marcin Marzencki, Farzad Khosrow-Khavar, Syed Ammar Zaidi, Carlo Menon and Bozena Kaminska

Abstract: Heart attack or myocardial infraction is the leading cause of deaths in the modern world. In order to increase survival chance of patients, action should be taken during the first hour from the onset of symptoms, which is most often impossible with the current technology. To this end, we propose a method of heart attack treatment based on low frequency diastolic timed vibrations. This method can be used in ambulatory setting by unspecialized personnel as it is noninvasive and safe for the patient. It is based on applying low frequency mechanical vibrations synchronized with the heart cycle of the subject along with application of thrombus dissolving drugs. We present an analysis of the proposed methodology and provide experimental results obtained with a prototype device.
Download

Paper Nr: 37
Title:

HEART RATE VARIABILITY ANALYSIS OF CHILDREN WITH REFRACTORY EPILEPSY BEFORE AND AFTER THE VAGUS NERVE STIMULATION

Authors:

Milica Milošević, Steven Vandeput, Sabine Van Huffel, Katrien Jansen and Lieven Lagae

Abstract: Vagus nerve stimulation (VNS) is a well-known therapeutic option for patients with refractory epilepsy who do not respond to adequate anti-epileptic drugs. Heart rate variability (HRV) is mediated by sympathetic and parasympathetic efferent activities which always interact towards the heart. Our goal was to describe the link between autonomic nervous system (ANS) and HRV. In 18 epileptic children, ECG data were obtained before and after implantation of the VNS. HRV was measured by linear and nonlinear parameters during 50 minute epochs during phase 2 of sleep and deep sleep. Results of the patients were compared with those of an age and sex matched control group. We were able to confirm that vagus nerve stimulation do not influence heart rate in children with refractory epilepsy. After the VNS implantation, there is a shift in sympathovagal balance towards sympathetic predominance in phase 2 of sleep (p=0.177) and also during deep sleep (p=0.035). This study suggests that left vagus nerve stimulation has little effect on heart rate variability as measured by nonlinear parameters.
Download

Paper Nr: 46
Title:

A COLLECTIVE BIOLOGICAL PROCESSING ALGORITHM FOR ECG SIGNALS

Authors:

Horia Mihail Teodorescu

Abstract: We establish and explore an analogy between hunting by packs of agents and signal processing. We present a version of adaptive ‘Hunting Swarm’ algorithm (HSA), apply it to ECG signals, and investigate the influence of the model parameters on the filtering of stationary and nonstationary noise. We show that results obtained with the HSA filter may outperform results obtained with several other filters.
Download

Paper Nr: 49
Title:

ECG ARTIFACT REMOVAL FROM SURFACE EMG SIGNALS BY COMBINING EMPIRICAL MODE DECOMPOSITION AND INDEPENDENT COMPONENT ANALYSIS

Authors:

Joachim Taelman, Bogdan Mijovic, Sabine Van Huffel, Stéphanie Devuyst and Thierry Dutoit

Abstract: The electrocardiography (ECG) artifact in surface electromyography (sEMG) is a major source of noise influencing the analyses. Moreover, in many cases the sEMG signal is the only available signal, making this removal more complicated. We compare the performance of two recently described single channel blind source separation methods with the commonly used template subtraction method on both simulations and real-life data. These two methods decompose a single channel recording into a multichannel representation before applying independent component analysis to these multichannel data. The decomposition methods are the wavelet decomposition and ensemble empirical mode decomposition (EEMD). The EEMD based single channel technique shows better performance compared to template subtraction and the wavelet based alternative for both high and low signal-to-artifact ratio and for simulated and real-life data, but at the expense of a higher computational load. We conclude that the EEMD based method has its potential in eliminating spike-like artifacts in electrophysiological signals.
Download

Paper Nr: 53
Title:

OPTIMISING CLASSIFIERS FOR THE DETECTION OF PHYSIOLOGICAL DETERIORATION IN PATIENT VITAL-SIGN DATA

Authors:

Sara Khalid, David A. Clifton, Lei Clifton and Lionel Tarassenko

Abstract: Hospital patient outcomes can be improved by the early identification of physiological deterioration. Automatic methods of detecting patient deterioration in vital-sign data typically attempt to identify deviations from assumed “normal” physiological condition. This paper investigates the use of a multi-class approach, in which “abnormal” physiology is modelled explicitly. The success of such a method relies on the accuracy of data annotations provided by clinical experts. We propose an approach to estimate class labels provided by clinicians, and refine those labels such they may be used to optimise a multi-class classifier for identifying patient deterioration. We demonstrate the effectiveness of the proposed methods using a large data-set acquired in a 24-bed step-down unit.
Download

Paper Nr: 64
Title:

AN ALGORITHM FOR THE DETECTION OF ATRIAL FIBRILLATION USING THE PULSE OXIMETRIC SIGNAL

Authors:

Giovanni Calcagnini, Michele Triventi, Federica Censi, Eugenio Mattei, Pietro Bartolini and Francesco Mele

Abstract: A method for the discrimination of atrial fibrillation and sinus rhythm from the pulse oximetric signal is presented. The method is based on the analysis of the ventricular rhythm irregularity, quantified by the Coefficient of Variation and the Shannon Entropy of the ventricular inter-beat intervals. A classifier based on the Mahalanobis distance is then applied. Sixty patients with an history of recurrent atrial fibrillation were studied. The method yielded a correct classification of 43 out of 43 patients with sinus rhythm, 14 out of 14 patients with atrial fibrillation, and 3 out of 4 patients with other arrhythmias.
Download

Paper Nr: 68
Title:

BIORESPONSE TO STEREOSCOPIC MOVIES PRESENTED VIA A HEAD-MOUNTED DISPLAY

Authors:

Takada Hiroki, Matsuura Yasuyuki, Fujikake Kazuhiro and Miyao Masaru

Abstract: Three-dimensional (3-D) television sets are already available in the market and are becoming increasingly popular among consumers. The 3-D movies they play, however, induce the negative sensations of asthenopia and motion sickness in some viewers. Visually induced motion sickness (VIMS) is caused by sensory conflict, i.e., a disagreement between vergence and visual accommodation during the viewing of stereoscopic images. VIMS can be analyzed both subjectively and physiologically. The objective of this study is to develop a method for detecting VIMS. We quantitatively measured head acceleration and body sway during viewer exposure to both a two-dimensional (2-D) image and a conventional three-dimensional (3-D) movie. The subjects wore head-mounted displays (HMDs) and maintained the Romberg posture for the first 60 s and a wide stance (midlines of the heels 20 cm apart) for the next 60 s. Head acceleration was measured using an active tracer at a sampling frequency of 50 Hz. Subjects completed the Simulator Sickness Questionnaire (SSQ) immediately afterwards. Statistical analysis was then applied to the SSQ subscores and to each index of stabilograms. Transfer function analysis indicated that the acceleration of the head in the anterior-posterior direction while watching a 3-D movie can affect lateral body sway, thereby causing VIMS.
Download

Paper Nr: 70
Title:

INCREASING RELIABILITY AND INFORMATION CONTENT OF PULSE OXIMETRIC SAHS SCREENING ALGORITHMS

Authors:

Nicole Gross, Jennifer Friedmann, Christophe Kunze, Wilhelm Stork, Daniel Sánchez Morillo, Antonio Leon Jimenez and Luis Felipe Crespo Foix

Abstract: About 3% of people suffer from sleep apnea-hypopnea syndrome SAHS. SAHS is a sleep associated respiratory disorder that negatively affects life quality and life expectancy. It is assumed that more than 80% of SAHS concerned are neither diagnosed nor therapied. Reliable and easy-handling SAHS screening systems are needed. Within this study, the reliability of pulse oximeter as a well-established, non-invasive medical device is examined for SAHS screening. Reliability of existing SAHS screening algorithms will be assessed. Hereby, the focus is on the influence of different desaturation detection strategies and the dependence on thresholds. Critiques on pulse oximetry as SAHS screening device will be responded. In this regard, guideline conform grey area integration in SAHS screening (concerning apnea-hypopnea index AHI between 5 and 15) is recommended. In particular, as by grey area integration an improvement of convenient SAHS screening algorithm reliabilities of about 7.3% in sensitivity and 8.7 % in specificity was achieved even in the most reliable tested algorithm. In a final step, room for improvement of screening results interpretation even without sleep medicine expert skills is indicated. In connection to this, possibilities of short-term frequency analysis of SpO2 data are demonstrated in its prospects for individualized SAHS screening quality.
Download

Paper Nr: 74
Title:

ON THE DETECTION OF BINARY CONCENTRATION-ENCODED UNICAST MOLECULAR COMMUNICATION IN NANONETWORKS

Authors:

Mohammad Upal Mahfuz, Dimitrios Makrakis and Hussein Mouftah

Abstract: Molecular communication is a new communication technique where transmitter and receiver communicate by transmitting molecules and correspondingly modulating their specific characteristics. Molecular communication is being considered as a new physical layer (PHY) option for a vast number of communicating nanomachines that form “nanonetworks.” Thus it has become a promising option for a large number of new applications, offering several benefits over conventional electromagnetic communications based on radio waves or optics at nanoscale dimension. Concentration-encoding is a simple and good technique to encode information with molecules. Incorrect detection of concentration-encoded signals makes molecular communication a real challenge. This paper has addressed sampling-based and energy-based detection approaches in detail for binary concentration-encoded molecular communication signals based on diffusion in fluidic media.
Download

Paper Nr: 75
Title:

A STUDY OF STOCHASTIC RESONANCE AS A MATHEMATICAL MODEL OF ELECTROGASTROGRAM

Authors:

Matsuura Yasuyuki, Miyao Masaru, Yokoyama Kiyoko and Takada Hiroki

Abstract: An electrogastrogram (EGG) is a recording of the electric activity of the stomach as measured on the abdominal surface. In this study, our goal is to obtain a mathematical model of an EGG obtained for a subject in the supine position. Initially, we applied the translation error in the Wayland algorithm to the EGG in order to measure the degree of determinism. However, we could not determine whether or not the mathematical model of the EGG could be defined on the basis of a chaotic process. The waveform of the electric potential in the interstitial cells of Cajal is similar to the graphs of the numerical solutions to the Van der Pol equation (VPE). We therefore added the VPE to a periodic function and random white noise was used to represent the intestinal motility and other biosignals, respectively. The EGG and numerical solutions were compared and evaluated on the basis of the translation error and the maximum Lyapunov exponent. The EGG was well described by the stochastic resonance in the stochastic differential equations.
Download

Paper Nr: 77
Title:

SIGNAL QUALITY ASSESSMENT FOR CAPACITIVE ECG MONITORING SYSTEMS USING BODY-SENSOR-IMPEDANCE

Authors:

Stephan Heuer, Sebastian Chiriac, Malte Kirst, Adnene Gharbi and Wilhelm Stork

Abstract: Contactless capacitive ECG measurement is an unobtrusive way of acquiring cardiovascular data. However, movement artifacts present a common problem with this technique. A means of assessing signal quality and confidence is therefore desirable. In this paper we present a capacitive ECG measurement system with an integrated module that constantly monitors the electrode-body-impedance. Moreover, we present a method to derive an artifact level signal from this electrode-body-impedance that can be used to estimate the signal quality of the capacitive ECG measurement. First results of measurements with this system are shown.
Download

Paper Nr: 78
Title:

AUTOMATIC DEPTH ELECTRODE LOCALIZATION IN INTRACRANIAL SPACE

Authors:

Janis Hofmanis, Valerie Louis-Dor, Olivier Caspary and Louis Maillard

Abstract: Localization and precise targeting of depth electrodes in the regions of the human brain is critical for accurate clinical diagnoses and treatment as well as for epileptical source localization and studies of in-vivo electrical propagation. By using magnetic resonance imaging (MRI) combined with computed tomography (CT), the authors present a method based on image processing and object recognition that improves electrode localization in different brain anatomies and matter. This method permits the quantified localization of electrode placements in cortex and white matter, and gives the precise position of each electrode, allowing a more detailed study of intra-cranial electrical stimulation, propagation and properties of conductivity related to the brain. Such methods can be extended to depth-scalp signal analysis using simultaneously registered SEEG and EEG measurements.
Download

Paper Nr: 79
Title:

PREPROCESSING IN MAGNETIC FIELD IMAGING DATA

Authors:

Dania Di Pietro Paolo, Tobias Toennis and Sergio Nicola Erne

Abstract: Magnetic Field Imaging (MFI) is a new method of diagnosis of increasing importance in cardiology. MFI records the magnetic fields (MF) of the electrical activity of the heart using many extremely sensitive sensors and displays them afterwards in a clinically applicable manner. Due to the relatively low signal to noise ratio (SNR) of the magnetic data, the recorded data are often averaged before analysis. We describe a standardized preprocessing method to be used before averaging MFI data with low SNR. The reported examples are data from 20 subjects out of a normal cohort examined at the Asklepios Klinik St. Georg.
Download

Paper Nr: 83
Title:

A NEW AND EFFICIENT VESSEL SEGMENTATION METHOD FROM COLOR RETINAL IMAGES

Authors:

Alauddin Bhuiyan, Ryo Kawasaki, Ecosse Lamoureux, Kotagiri Ramamohanarao and Tien Y. Wong

Abstract: Retinal blood vessel changes (e.g., vessel caliber) are important indicators for earlier diagnosis of cardiovascular diseases. To quantify the changes automatically, a reliable vessel detection is essential. However, blood vessel detection in retinal image is complicated by a huge variation in a number of factors such as local contrast, vessel width and vessel central reflex. In this paper, we propose a new technique to detect retinal blood vessels which is able to address these issues. The core of the technique is a new vessel edge selection method which combines the method of finding edge pattern and edge profiling techniques. Experimental results show that 92.40% success rate in the identification of vessel start-points and 88.73% success rate in tracking the major vessels.
Download

Paper Nr: 90
Title:

A LEARNING APPROACH TO IDENTIFICATION OF NONLINEAR PHYSIOLOGICAL SYSTEMS USING WIENER MODELS

Authors:

Xingjian Jing, Natalia Angarita-Jaimes, David Simpson, Robert Allen and Philip Newland

Abstract: The Wiener model is a natural description of many physiological systems. Although there have been a number of algorithms proposed for the identification of Wiener models, most of the existing approaches were developed under some restrictive assumptions of the system such as a white noise input, part or full invertibility of the nonlinearity, or known nonlinearity. In this study a new recursive algorithm based on Lyapunov stability theory is presented for the identification of Wiener systems with unknown and noninvertible nonlinearity and noisy data. The new algorithm can guarantee global convergence of the estimation error to a small region around zero and is as easy to implement as the well-known back propagation algorithm. Theoretical analysis and example studies show the effectiveness and advantages of the proposed method compared with the earlier approaches.
Download

Paper Nr: 91
Title:

A PULSE WAVEFORM DATA DECOMPOSITION BASED ON MULTI COMPONENT CURVE FIT COMPARED WITH SECOND DERIVATIVE PHOTOPLETHYSMOGRAPHY AND PHASE PLANE PLOT

Authors:

M. Huotari, K. Maatta and J. Kostamovaara

Abstract: With a new photoplethysmographic (PPG) device we have been attending to photoplethysmographic signals of different ages for signal decomposition purpose. Because PPG is a non-invasive, and easily attachable measurement technique both suitable in health care applications we concentrated on its comprehensive signal analysis and waveform interpretation. By means of PPG it is easy to capture data for further analysis. In the world cardiovascular diseases are the frequent cause of death that’s why we are concern on cardiovascular diseases. The main cause of incidents can be high arterial stiffness which is symptomless and increases the risk as a function of age causing cardiovascular diseases. Arteries stiffen normally as a consequence of age, but also because of insalubrious mores and many diseases. Normal age related stiffness occurs when the elastic fibers within the arterial walls begin to weaken due to age, but diseases as arteriosclerosis accelerate this process. However, we believe that it is possible to prevent arterial stiffening if detected early enough. For this reason we have derived indexes to indicate a possible arterial stiffness value.
Download

Paper Nr: 92
Title:

MEASURING P-WAVE MORPHOLOGICAL VARIABILITY FOR AF-PRONE PATIENTS IDENTIFICATION

Authors:

Valeria Villani, Antonio Fasano, Luca Vollero, Federica Censi and Giuseppe Boriani

Abstract: Atrial fibrillation is the most common arrhythmia encountered in clinical practice. Abnormal P-waves have been observed in patients prone to AF and the analysis of P-waves from surface electrocardiogram has been extensively used to identify patients prone to atrial arrhythmias. Measuring the temporal variability of Pwaves, i.e., the variation over time of morphological characteristics of single P-waves, may represent a useful method for characterizing and predicting AF cases. In this paper, we propose a method for the statistical analysis of P-waves variability. It is based on the evaluation of the empirical distribution function of differences energy among normalized P-waves. The proposed method seems promising for capturing atrial anomalies and identifying patients prone to AF.
Download

Paper Nr: 95
Title:

COMPARING DRY ELECTRODE MATERIALS FOR LONG-TERM ECG MONITORING - Evaluation of a New Method

Authors:

Silvester Fuhrhop, Stefan Lamparth, Tobias Seemann, Wilhelm Stork, Stephan Heuer and Malte Kirst

Abstract: Studies prove, that ECG monitoring over longer periods of time can significantly improve the diagnosis of heart failure. Dry electrodes are a valuable alternative to conventional wet electrodes, if the monitoring period is longer than one week, because they are more stable and cause less skin irritations. One drawback however is the minor signal quality compared to wet electrodes. As there are different electrode materials for dry electrodes available, a qualitative comparison concerning signal quality is carried out in this work. The signal quality of well known dry-electrode-materials used in fitness-heart belts are compared with metal dry electrodes. All electrodes are assembled in an identical mechanical electrode-setup. A parcours-based field trial with 13 subjects is carried out in two stages. In order to estimate the reproducibility of the parcours, the first stage is used to benchmark the differences in signal quality of consecutive measurements not originating from the electrode material. In the second stage however, the actual measurement takes place. The accuracy of the method is presented and the comparison results are discussed.
Download

Paper Nr: 97
Title:

FETAL CARDIAC BYPASS ANALYSIS BY MEANS OF CORRELATION DIMENSIONS

Authors:

G. D'Avenio, C. Daniele and M. Grigioni

Abstract: During in vivo experiments of fetal cardiac surgery performed in sheep, physiological signals were recorded, and subsequently analyzed. In order to characterize their complexity, the fractal dimension was calculated. The adopted model of dimension estimation allowed for a possible multifractal nature of the signals, by considering two distinct fractal dimensions ѵ1, ѵ 2 at different length scales. A comparison was also carried out with an alternative measure of system complexity, Approximate Entropy (ApEn). The results of the analysis suggest that fractal dimension may be a useful indicator of the cardiac stress and, ultimately, of the quality of the support delivered during the operation.
Download

Paper Nr: 103
Title:

A NEW WATERMARKING SCHEME BASED ON MULTI-WATERMAKS EMBEDDING IN MULTI-DOMAINS USING FUZZY C-MEANS TECHNIQUE

Authors:

Sameh Oueslati, Adnane Cherif and Bassel Solaiman

Abstract: Watermarking is used for the protection of intellectual property, data integrity, and data authentication. This paper proposes a novel method for image watermarking based on embedding multiples watermarks in different domains of the image representation (spatial and DCT domains), without any distortion of the watermarked image. In the spatial domain, the processing method is based on study of segmentation by fuzzy c-means clustering method that outputs the zones of watermark embedding and respectively the associated appropriate embedding gain factors. However in the DCT domain a proper choice of the DCT coefficients based on the quantization JPEG table in the middle frequencies band is carried out. Several watermarks were embedded in these two domains in order to take advantage of the spatial domain robustness against different asynchronous attacks, associated to the DCT domain robustness against jpeg compression and some other signal processing distortions. Experimental results show that the proposed method is robust against a large set of synchronous and asynchronous image attacks such as filtering, lossy compression, cropping and rotation attack.

Paper Nr: 107
Title:

NEURAL CLASSIFIER FOR DETECTION AND CLASSIFICATION OF SPIKES AND SHARP WAVES

Authors:

Fernando Mendes de Azevedo, Geovani Rodrigo Scolaro and Christine Fredel Boos

Abstract: In this article is discussed the application of a hybrid approach that uses the Wavelet Transform and Artificial Neural Networks in detection and recognition of epileptiform events in EEG signals. It is presented the methodology used to develop a Neural Classifier as well as the experiments and its results through the Neural Networks and Wavelet Transform. The developed Neural Classifier showed good results in the classification of Epileptiform events with and without pre-processing achieving sensitive of 97.14%, specificity of 94.55% and accuracy of 96.14%, suggesting the high sample rate of the EEG signals contributed to achieve these values.
Download

Paper Nr: 108
Title:

MORPHOLOGICAL ANALYSIS OF ACCELERATION SIGNALS IN CROSS-COUNTRY SKIING - Information Extraction and Technique Transitions Detection

Authors:

Håvard Myklebust, Neuza Nunes, Jostein Hallén and Hugo Gamboa

Abstract: Aims: Experience morphology of acceleration signals, extract useful information and classify time periods into defined techniques during cross-country skiing. Method: Three Norwegian cross-country skiers ski skated one lap in the 2011 world championship sprint track as fast as possible with 5 accelerometers attached to their body and equipment. Algorithms for detecting ski/pole hits and leaves and computing specific ski parameters like cycle times (CT), poling/pushing times (PT), recovery times (RT), symmetry between left and right side and technique transition times were developed based on thresholds and validated against video. Results: In stable and repeated techniques, pole hits/leaves and ski leaves were detected 99% correctly, while ski hits were more difficult to detect (77%). From these hit and leave values CT, PT, RT, symmetry and technique transitions were successfully calculated. Conclusion: Accelerometers can definitely contribute to biomechanical research in cross-country skiing and studies combining force, position and accelerometer data will probably be seen more frequently in the future.
Download

Paper Nr: 115
Title:

TWO LINK COMPLIANT ROBOT MANIPULATOR FOR HUMAN ROBOT COLLISION SAFETY

Authors:

Muhammad Rehan Ahmed and Ivan Kalaykov

Abstract: For successful human robot interaction (HRI), collision safety as well as position accuracy are equally important. Robot is required to demonstrate safe sharing of work space with humans and to exhibit adaptable compliant behavior that comply with interaction forces generated upon contact. We present an approach for acquiring reconfigurable compliance using semi-active actuation mechanism, where compliance is achieved by controlling the viscous properties of magneto-rheological (MR) fluid. In this paper, we have discussed three essential modes of motions required for safe physical HRI. Then, we have shown collision safety of our robot based on static and dynamic collision testing in different motion modes. Finally, experimental results validate the significance of our proposed approach for human robot collision safety and high position accuracy.
Download

Paper Nr: 126
Title:

CONSTRUCTING A SYSTEM TO EVALUATE EFFECTS OF SUPPORT TAPING FOR ANKLE INVERSION SPRAIN USING 3-D MOTION AND PLANTER PRESSURE

Authors:

Jun Akazawa, Takaharu Ikeuchi, Takemasa Okamoto and Ryuhei Okuno

Abstract: In the field of sports science,support taping for ankle inversion sprain has often been used. The motion of ankle joint would be limited with support taping for ankle inversion. In order to clarify the effects of the ankle taping and to examine characteristics of the taping, we had constructed a system to measure the distance between the metatarsus first head and the floor with 3D motion analysis system, and to measure the planter pressure patterns during the ankle inversion with pressure monitoring system. When the eight subjects were instructed to inverse their ankles as much as possible with and without taping, there was a difference in the distances between taping and no taping.
Download

Paper Nr: 127
Title:

NOVEL APPROACH TO CHEST IMPEDANCE SIGNAL ANALYSIS

Authors:

Algimantas Krisciukaitis, Andrius Macas, Renata Simoliuniene, Robertas Petrolis and Zita Bertasiene

Abstract: New wave of development of more informative and reliable diagnostic methods substituting classical Impedance Cardiography introduced by Sramek in the 1960's was inspired by rapid development of IT based devices in medicine. We illustrate approaches of multivariate analysis of chest impedance signals in aim to reveal parameters reflecting detail pattern of functions of cardiovascular system.
Download

Paper Nr: 130
Title:

AUTOMATIC REMOVAL OF SPARSE ARTIFACTS IN ELECTROENCEPHALOGRAM

Authors:

Petr Tichavský, Miroslav Zima and Vladimir Krajča

Abstract: In this paper we propose a method to identify and remove artifacts, that have a relatively short duration, from complex EEG data. The method is based on the application of an ICA algorithm to three non-overlapping partitions of a given data, selection of sparse independent components, removal of the component, and the combination of three resultant signal reconstructions in one final reconstruction. The method can be further enhanced by applying wavelet de-noising of the separated artifact components.
Download

Paper Nr: 131
Title:

VISUALIZATION IN SPORTING CONTEXTS - The Team Scenario

Authors:

Aqeel H. Kazmi, Michael J. O’Grady and Gregory M. P. O’Hare

Abstract: Wearable sensor systems require an interactive and communicative interface for the user to interpret data in a meaningful way. The development of adaptive personalization features in a visualization tool for such systems can convey a more meaningful picture to the user of the system. In this paper, a visualization tool called Visualization in Team Scenarios (VTS), which can be used by a coach to monitor an athlete’s physiological parameters, is presented. The VTS has been implemented with a wearable sensor system that can monitor players’ performance in a game in a seamless and transparent manner. Using the VTS, a coach is able to analyze the physiological data of athletes generated using select wearable sensors, and subsequently analyse the results to personalize training schedules thus improving the performance of the players.
Download

Paper Nr: 147
Title:

AN APPROACH OF REDUCING MEASURE TIME OF NONINVASIVE THERMOMETER - Application of Curve-fitting Method and Autoregressive Model for Reducing the Measure Time of Dual-heat-flux Thermometer

Authors:

S. Y. Sim, H. J. Baek, G. S. Chung and K. S. Park

Abstract: Newly developed dual-heat-flux thermometer is expected to be useful in measuring core body temperature noninvasively. However, as it takes more than 30 min to measure, the additional process is needed to reduce the measure time. In this study, we made a dual-heat-flux thermometer to verify its performance and obtained an hour-long data from three subjects. Dual-heat-flux thermometer estimated the core body temperature very well in all subjects. In addition, least squares curve-fitting method predicted deep body temperature well with within 100 sec data. Autoregressive model with 10 sec data also seemed to be suitable method for shortening measure time of dual-heat-flux thermometer.
Download

Paper Nr: 162
Title:

MUSCLE ACTIVATION THRESHOLDS BEFORE AND AFTER TOTAL KNEE ARTHROPLASTY - Protocol of a Randomized Comparison of Minimally Invasive vs. Standard Approach

Authors:

Carlos J. Marques, Hugo Gamboa, Frank Lampe, João Barreiros and Jan Cabri

Abstract: After total knee arthroplasty (TKA) patients often ask when they can resume car driving. This question was the aim of some studies in the past, however it is not clear whether minimally invasive surgery (MIS) for total knee replacement has benefits in terms of faster recovering times. With the present study protocol the effects of two surgery techniques for TKA (MIS vs. standard approach) on motor performance parameters will be tested during the performance of an emergency brake in a car simulator. The brake response time components and the muscle activation thresholds of four muscles involved in the task will be the outcomes of the study.
Download