BIOSIGNALS 2010 Abstracts

Full Papers Short Papers
Full Papers
Paper Nr: 32
Title:

Probabilistic Patient Monitoring Using Extreme Value Theory
Authors:

David Clifton, Lionel Tarassenko and Samuel Hugueny
Abstract: Conventional patient monitoring is performed by generating alarms when vital signs exceed pre-determined thresholds, but the false-alarm rate of such monitors in hospitals is so high that alarms are typically ignored. We propose a principled, probabilistic method for combining vital signs into a multivariate model of patient state, using extreme value theory (EVT) to generate robust alarms if a patient's vital signs are deemed to have become sufficiently "extreme". Our proposed formulation operates many orders of magnitude faster than existing methods, allowing on-line learning of models, leading ultimately to patient-specific monitoring.
Download
Paper Nr: 35
Title:

Feature relevance assessment in automatic inter-patient heart beat classification
Authors:

Gael de Lannoy, Damien Francois, M. Verleysen and J. Delbeke
Abstract: Long-term ECG recordings are often required for the monitoring of the cardiac function in clinical applications. Due to the high number of beats to evaluate, inter-patient computer-aided heart beat classification is of great importance for physicians. The main difficulty is the extraction of discriminative features from the heart beat time series. The objective of this work is the assessment of the relevance of feature sets previously proposed in the literature. For this purpose, inter-patient classification of heart beats following AAMI guidelines is investigated. The class unbalance is taken into account by using a support vector machine (SVM) classifier that integrates distinct weights for the classes. The performances of the SVM model with an appropriate selection of features are better than those of previously reported inter-patient classification models. These results show that the choice of the features is of major importance, and that some usual feature sets do not serve the classification performances. In addition, the results drop significantly when the class unbalance is not taken into account, which shows that this issue must be addressed to grasp the importance of the pathological cases.
Download
Paper Nr: 36
Title:

RESONANCES IN THE CARDIOVASCULAR SYSTEM: INVESTIGATION AND CLINICAL APPLICATIONS
Authors:

Evgeny G. Vaschillo, Bronya Vaschillo, Jennifer Buckman, Marsha Bates and Robert J. Pandina
Abstract: The baroreflex, as a control system with negative feedback, is a mechanism that buffers changes in blood pressure (BP), thereby precluding strong, abrupt shifts in arterial pressure. As a closed-loop control system with delays, the baroreflex possesses resonance features at frequencies of about 0.1 and 0.03 Hz. These resonance frequencies correspond to a ~5-s delay in the BP response to changes in heart rate (HR) (HR baroreflex closed-loop) and a ~15-s delay in the vascular tone (VT) response to changes in BP (VT baroreflex closed-loop). Thus, whereas a single impact on the cardiovascular system (CVS) elicits a HR, BP, and VT oscillatory response that fades over time, 0.1 or 0.03 Hz rhythmical stimulation of the CVS produces steady HR, BP, and VT oscillations with significantly higher amplitudes comparing to stimulation at other frequencies. Resonances in the baroreflex system are essential for the maintenance of optimal health by keeping autonomic regulation active via HR, BP, and VT variability, providing adaptive responses to internal and external stimuli, and buffering stress and emotional reactivity via inhibitory effect in the brain. This study investigates the phenomenon of resonances in the CVS and the ability to employ these resonances for clinical applications.
Download
Paper Nr: 54
Title:

COMBINING TEMPORAL AND FREQUENCY BASED PREDICTION FOR EEG SIGNALS
Authors:

Padma P. Paul, Howard Leung, David A. Peterson, Howard Poizner and Terrence J. Sejnowski
Abstract: This paper presents a novel approach for electroencephalogram (EEG) signal prediction. It combines temporal and frequency based prediction to achieve a good final prediction result. Artificial neural networks are used as the predictive model for signals both in the temporal and frequency domain. In frequency based prediction, the amplitude and the phase of the frequency response are predicted separately. Experiments were conducted on the prediction of EEG data recorded from 13 subjects. Eight performance measures were used to evaluate the performance of our proposed method. Experiment results show that the proposed combined prediction method gives the overall best performance compared with the temporal based prediction alone and the frequency based prediction alone.
Download
Paper Nr: 55
Title:

RELEVANCE AND LOCI OF ODORANT FEATURES IN THE RAT OLFACTORY BULB - Statistical Methods for Understanding Olfactory Codes in Glomerular Images
Authors:

Benjamin Auffarth, Benjamin Auffarth, Agustin Gutierrez and Santiago Marco
Abstract: The relationship between physicochemical properties of odor molecules and perceived odor quality is arguably one of the most important issues in olfaction and the rules governing this relationship remain unknown. Any given odor molecule will stimulate more than one type of receptor in the nose, perhaps hundreds, and this stimulation reflects itself in the neural code of the olfactory nervous system. We present a method to investigate neural coding at the glomerular level of the olfactory bulb, the first relay for olfactory processing in the brain. Our results give insights into localization of coding sites, relevance of odorant properties for information processing, and the size of coding zones.
Download
Paper Nr: 63
Title:

PITCH-ASYNCHRONOUS GLOTTAL INVERSE FILTERING OF NORMAL AND PATHOLOGICAL VOICES BASED ON HOMOMORPHIC PREDICTION
Authors:

Rubén Fraile, Malte Kob, Juana M. Gutierrez, Nicolás Sáenz-Lechón, Juan I. Godino-Llorente and Víctor Osma-Ruiz
Abstract: Inverse filtering of speech signals for the separation of vocal tract and glottal source effects has a wide variety of potential applications, including the assessment of glottis-related aspects of voice function. Among all existing approaches to inverse filtering, this paper focuses on homomorphic prediction. While not favoured much by researchers in recent literature, such an approach offers two advantages over others: it does not require previous estimation of the fundamental frequency and it does not rely on any assumptions about the spectral enevelope of the glottal signal. The performance of homomorphic prediction is herein assessed and compared to that of an adaptive inverse filtering method making use of synthetic voices produced with a biomechanical voice production model. The reported results indicate that the performance of inverse filtering based on homomorphic prediction is within the range of that of adaptive inverse filtering and, at the same time, it has a better behaviour when the spectral envelope of the glottal signal does not suit an all-pole model of predefined order.
Download
Paper Nr: 68
Title:

Particle Swarm Optimisation Aided Multiuser Transmission Schemes for MIMO Communication
Authors:

Wang Yao, Sheng Chen and Lajos Hanzo
Abstract: Bio-inspired computational methods have found wide-ranging applications in signal processing and other walks of engineering. In this contribution, particle swarm optimisation (PSO) is invoked for designing optimal multiuser transmission (MUT) schemes for multiple-input multiple-output communication. Specifically, we consider the minimum bit-error-rate (MBER) linear MUT using PSO and we design a PSO aided MBER generalised vector precoding for nonlinear MUT. These PSO aided MUT techniques compare favourably with the state-of-the-art conventional schemes, in terms of performance and complexity.
Download
Paper Nr: 73
Title:

Modeling Complexity of Physiological Time Series In-silico
Authors:

Konstantin Mischaikow, Jesse Berwald, Tomas Gedeon and Konstantin Mischaikow
Abstract: A free-running physiological system produces time series with complexity which has been correlated to the robustness and health of the system. The essential tool to study the link between the structure of the system and the complexity of the series it produces is a mathematical model that is capable of reproducing the statistical signatures of a physiological time series. We construct a model based on the neural structure of the hippocam- pus that reproduces detrended fluctuations and multiscale entropy complexity signatures of physiological time series. We study the dependence of these signatures on the length of the series and on the initial data.
Download
Paper Nr: 75
Title:

"Rain Fall" Particle Model for Shape Recovery and Image Segmentation
Authors:

Shahram Payandeh and Wen Shi
Abstract: This paper studies the problem of shape recovery and image segmentation with examples related to medical imaging. Our purpose is to explore an alternative physics based image segmentation model in comparison with parametric intensive methods such as active contour or level set approaches. The proposed model can offer a more computational efficient approach. As an early attempt, a novel segmentation method based on physically motivated particle system is presented, analyzed and integrated for 2D and 3D applications. Different from previous particle based segmentation method, our proposed approach is governed physically by fluid dynamic model. Additionally a novel "rain fall" model is presented as an alternative paradigm for shape reconstruction and image segmentation when working with complex 2D and 3D medical images. In this paper, an overview of fluid mechanical model and fluid particle simulation process is presented as well. Segmentation results on 2D images and shape recovery of 3D images are presented followed by discussions and conclusions.
Download
Paper Nr: 89
Title:

Tactile texture discrimination in the robot-rat Psikharpax
Authors:

Jean-Arcady Meyer, Steve N'Guyen, Patrick Pirim and Jean-Arcady Meyer
Abstract: We endowed a whiskered robot with a simple algorithm allowing to discriminate textures. Its efficiency and robustness have been demonstrated using both a fixed head and a mobile platform. Comparatively to previous similar approaches, this system affords greater behavioral capacities and proves to be able to complement or supply vision in simple navigation tasks. The corresponding results suggest that the length and number of the whiskers involved play a role in texture discrimination. They also suggest that two hypotheses that are currently considered as mutually exclusive to explain texture recognition in rats - i.e., the ``kinetic signature hypothesis'' and the ``resonance hypothesis'' - may be, in fact, complementary.
Download
Paper Nr: 99
Title:

HEART RATE VARIABILITY MEASUREMENT USING THE SECOND DERIVATIVE PHOTOPLETHYSMOGRAM
Authors:

Mirjam Jonkman, Mohamed Elgendi, Friso De Boer and Mirjam Jonkman
Abstract: Heart-rate monitoring is a basic measure for cardiovascular functionality assessment. The electrocardiogram (ECG) and Holter monitoring devices are accurate, but their use in the field is limited. Photoplethysmography is an optical technique that has been developed for experimental use in vascular disease. Because of its non-invasive, safe, and easy-to-use properties, it is considered a promising tool that may replace some of the current traditional cardiovascular diagnostic tools. A useful algorithm for a wave detection in the second derivative plethysmogram (SDPTG) is introduced for heart–rate monitoring. The performance of the proposed method was tested on 27 records measured at rest and after exercise. Statistical HRV measures can be calculated using the a-a interval of the SDPTG.
Download
Paper Nr: 100
Title:

STUDY OF EFFECTIVE CONNECTIVITY FOR FACE PERCEPTION IN HEALTHY SUBJECTS AND PARKINSON'S DISEASE
Authors:

Elvis Silva, João Sato, Ellison Cardoso, Edson Junior and Gabriela Castellano
Abstract: Facial perception is a fundamental task in our daily life and plays a critical role in social interactions. Evidence from neuropsychological, neurophysiologic, and functional imaging studies indicated that face perception is mediated by a specialized system in the human brain. We investigated the neural connectivity induced by face presentation with different emotional valences in Parkinson's disease (PD) patients and a control group of healthy, drug-free volunteers, using event-related fMRI in a parametric design. In this study, we focused on applying Dynamic Causal Modelling (DCM), an approach that allows the assessment of effective connectivity within cortical networks (Friston et al. 2003), to the study of effective connectivity between maximally activated brain regions in response to passive viewing of facial stimuli. A connectivity model was built based on the literature and in our fMRI analyses, which included the fusiform gyrus, anterior cingulate gyrus, dorsolateral prefrontal cortex (DLPFC) and dorsomedial prefrontal cortex (DMPFC). The results showed differences in connectivity between the PD group and the control group. We found that the effective couplings among DLPFC/DMPFC and FG, DLPFC/DMPFC and ACG, were higher in PD patients than healthy subjects, while the effective coupling among FG and ACG was lower in PD patients.
Download
Paper Nr: 112
Title:

Evaluation system for monitoring of vital parameters and active body climate control
Authors:

K. D. Mueller-Glaser and Adnene Gharbi
Abstract: This work presents a textile integrated evaluation system for active body climate control. The evaluation system registers several vital parameters of the user (skin temperature, skin relative humidity, heart rate, breathing rate and 3D acceleration data), its current subjective feedback and some surrounding parameters (temperature and relative humidity) and thus automatically controls the air ventilation level inside a cooling vest. For the climate control, a regulation algorithm influencing the body heat exchange processes and leading to thermal comfort at different workloads and different surrounding conditions is heuristically designed. In addition, a field study is conducted. This study involves 11 test persons and aims at validating the sensor data of the evaluation system and determining the energy expenditure of the body from the sensor data by analyzing the correlation between these data and the reference data of a spirometer. Besides, a verification of the suitability of the evaluation system for daily use and a validation of the implemented regulation algorithm is conducted.
Download
Paper Nr: 115
Title:

BIOMETRIC AUTHENTICATION USING BRAIN RESPONSES TO VISUAL STIMULI
Authors:

André Zúquete, Bruno Quintela and J. S. Cunha
Abstract: This paper studies the suitability of brain activity, namely electroencephalogram signals, as raw material for conducting biometric authentication of individuals. Brain responses were extracted with visual stimulation, leading to biological brain responses known as Visual Evoked Potentials. We evaluated a novel method, using only 8 occipital electrodes and the energy of differential EEG signals, to extract information about the subjects for further use as their biometric features. To classify the features obtained from each individual, we used a one-class classifier per subject and we tested four types of classifiers: K-Nearest Neighbor, Support Vector Data Description and two other classifiers resulting from the combination of the two ones previously mentioned. After testing these four classifiers with features of 70 subjects, the results showed that visual evoked potentials are suitable for an accurate biometric authentication.
Download
Short Papers
Paper Nr: 2
Title:

ENTROPIC ANALYSIS AND SYNTHESIS OF BIOSIGNAL COMPLEXITY
Authors:

Tuan Pham
Abstract: Analysis of complexity of biological time-series data is investigated to gain knowledge about the biosignal predictability. Using modern biological data such as mass spectral, this complexity information can be utilized to identify novel biomarkers for drug discovery, early disease detection and therapeutic treatment. To enhance the complexity analysis, a probabilistic fusion scheme, which is an alternative to the assumption of the independence of probabilistic models, is applied to synthesize the information given by different entropy methods.
Download
Paper Nr: 6
Title:

AUTOMATIC SEGMENTATION OF EMBRYONIC HEART IN TIME-LAPSE FLUORESCENCE MICROSCOPY IMAGE
Authors:

Petra Krämer, Céline Paloc, Olaia Holgado, Juan Mari Virto , Diana Wald, Fernando Boto, Ainhoa Letamendia, Fabien Bessy, Carles Callol and Izaskun Ibarbia

Abstract: Embryos of animal models are becoming widely used to study cardiac development and genetics. However, the analysis of the embryonic heart is still mostly done manually. This is a very laborious and expensive task as each embryo has to be inspected visually by a biologist. We therefore propose to automatically segment the embryonic heart from high-speed fluorescence microscopy image sequences, allowing morphological and functional quantitative features of cardiac activity to be extracted. Several methods are presented and compared within a large range of images, varying in quality, acquisition parameters, and embryos position. Although manual control and visual assessment would still be necessary, the best of our methods has the potential to drastically reduce biologist workload by automating manual segmentation.
Download
Paper Nr: 9
Title:

MIMO INSTANTANEOUS BLIND IDENTIFICATION BASED ON THIRD-ORDER CUMULANT
Authors:

Shen Xizhong
Abstract: This paper presents a new MIMO instantaneous blind identification algorithm based on third-order temporal property. Third-order temporal structure is reformulated in a particular way such that each column of the unknown mixing matrix satisfies a system of nonlinear multivariate homogeneous polynomial equations. The nonlinear system is solved by improved steepest descent method. We construct a general goal of the nonlinear system and convert the nonlinear problem into an optimal problem. The optimal solutions are obtained one by one by adding a penalty item to the general goal, which is Gaussian function characterized with valley-filled feature. Our algorithm allows estimating the mixing matrix for scenarios with 3 sources and 2 sensors, etc. Finally, simulations and comparisons show its effectiveness.
Download
Paper Nr: 12
Title:

COMPARISON BETWEEN LASER DOPPLER FLOWMETRY SIGNALS RECORDED IN GLABROUS AND NON GLABROUS SKIN: TIME AND FREQUENCY ANALYSES
Authors:

Edite Figueiras, Abraham Pierre, Requicha F. Luis and Humeau Anne
Abstract: Skin microvascular properties vary with anatomical zones. Thus, glabrous skin found in fingers, toes, nail beds, hand palms and feet soles has a high density of arteriovenous anastomoses (AVAs). In contrast, skin found in sites such as ventral face of the forearms do not possess AVAs and therefore microvascular blood flow in this non glabrous skin is different. We herein propose to analyse laser Doppler flowmetry (LDF) signals that reflect skin microvascular perfusion, in two different sites of healthy subjects: hand (glabrous skin) and ventral face of the forearm (non glabrous skin). The signal analysis is performed both in the time and in the frequency domains. Our results show that the mean amplitude of LDF signals recorded in the hand is generally higher than in the forearm. Moreover, the signal fluctuations observed in the hand are much higher than the ones observed in the forearm. Our work also shows that the power spectrum of LDF signals recorded in hand and forearm can be different. They both may possess characteristics of fractal processes but these characteristics may be different for the two anatomical sites.
Download
Paper Nr: 13
Title:

ENHANCED METHOD FOR ROBUST MOOD EXTRACTION FROM SKIN CONDUCTANCE
Authors:

Gert-Jan de Vries and Marjolein van der Zwaag
Abstract: One of the key challenges in affective computing is the interpretation of physiological signals into affect. Mood, as a subclass of affect, is known to be reflected in skin conductance. While most reports concern strictly controlled laboratory settings, daily life situations pose more challenges in interpreting physiology because more bodily and cognitive processes that influence skin conductivity are involved; for example temperature regulation or physical and mental activity. Existing techniques to reduce the effects of these processes in order to extract mood from skin conductance are rather crude and leave room for improvement. We introduce a more sophisticated method based on skin conductance response subtraction that provides better resemblance with mood. Validation of our method, using comparison with two alternative methods, shows our method excels in differentiation between positive and negative moods from skin conductance. Our method thereby enhances mood extraction from skin conductance, thus improving robustness of mood measurements.
Download
Paper Nr: 21
Title:

MODEL DEVELOPMENT FOR PROPOFOL AND REMIFENTANIL MANAGEMENT DURING ICU ANESTHESIA

Authors:

Ramona Hodrea, Sandra Iulianetti, Robin De Keyser and Clara Ionescu
Abstract: This paper presents the development of a MISO (multiple-input single-output) patient model for sedation and analgesia components used in ICU. The two inputs are Propofol and Remifentanil and the output is the Bispectral Index. The MISO model consists of two well-known PK-PD models for Propofol and Remifentanil, and an interaction model which describes the synergistic effect of these two drugs on the Bispectral Index. The interaction model parameters were identified using a nonlinear least squares method. Data collected during clinical trials in ICU at Ghent University Hospital have been used for model development. The final purpose is to use this model for prediction in a model based predictive control strategy.
Download
Paper Nr: 27
Title:

Simultaneous Autofocusing and Contouring of Human Ovocytes and Zygotes for In Vitro Fertilization
Authors:

Alessandro Giusti, Alessandro Giusti, Giorgio Corani, Luca Gambardella and Cristina Magli
Abstract: Observation of ovocytes and zygotes plays an important role in In Vitro Fertilization procedures, and is usually perfomed by means of a microscope equipped with Hoffman Modulation Contrast optics, which produces images with a complex, side-lit, 3D-like appearance. Our algorithm operates on a Z-stack of such images taken at different focal planes, and simultaneously identifies: a) a repeatable, meaningful focal plane corresponding to the cell's equator line, and b) the external contour of the cell. As the cell is a thick stucture with respect to the microscope depth of field, the two problems are nontrivial and strongly related. Our algorithm is also robust to other structures, clutter and artefacts affecting the images and lying at varying focal planes. We point out implementation details and applications of our technique, which has been integrated in the workflow of an IVF laboratory.
Download
Paper Nr: 30
Title:

DYNAMIC IMAGE SEGMENTATION SYSTEM WITH MULTI-SCALING SYSTEM FOR GRAY SCALE IMAGE
Authors:

Ken'ichi Fujimoto, Mio Musashi and Tetsuya Yoshinaga
Abstract: In this paper, we describe an image segmentation technique for a gray scale image by utilizing the nonlinear dynamics of two respective discrete-time dynamical systems. The authors have proposed a discrete-time dynamical system that consists of a global inhibitor and chaotic neurons that can generate oscillatory responses. By utilizing oscillatory responses, our system can perform dynamic image segmentation, which denotes the function that segments image regions and concurrently exhibits segmented images in time series, for a binary image. In order that our system can work well for a gray scale image, we introduce an extender as a pre-processing unit for our system. The extender also consists of a discrete-time dynamical system and can find an image region composed of pixels with different gray levels by multi-scaling gray levels of all pixels. In addition, it can compute the proximity between pixels based on their multi-scaled gray levels. Found values of the proximity can be used for determining couplings of chaotic neurons in our system. We demonstrated that our dynamic image segmentation system with the multi-scaling system works well for a gray scale image.
Download
Paper Nr: 37
Title:

HUMAN BREAST SHAPE ANALYSIS USING PCA
Authors:

Giovanni Gallo, Giuseppe C. Guarnera and Giuseppe Catanuto
Abstract: This paper introduces a parametric space to describe the shape of human breasts. The parameter space has been obtained from a sample of about 40 patient’s MRI taken in prone position. The data have been cleaned from noise and disturbances and has been dimensionally reduced using Principal Component Analysis. If two references relative to extremal shapes (one of a reconstructed breast and one of a severely aged breast) are taken, all the other shapes span a continuum space that provides an objective way to classify and describe the variability observed in the common clinical practice.
Download
Paper Nr: 42
Title:

Respiratory Information in Arterial Oxygen Saturation Measurement
Authors:

Yue-Der Lin
Abstract: Pulse oximeter has become a standard in intensive and critical care units for the monitoring of oxygen support from respiratory system since 1990’s. The multi-wavelength photoplethysmography (PPG) technique is now utilized for the measurement of arterial oxygen saturation by pulse oximeter (SpO2). This research utilized multi-channel autoregressive (AR) spectral estimation method for the coherence analysis between the respiratory signal and the PPG signal derived from pulse oximeter. Five healthy male subjects participated in this research with signals being measured at different respiratory status. The results demonstrate high coherence between respiration and the PPG signal from pulse oximeter, and the coherence disappears in breath-holding experiments. The results demonstrate that the respiratory status can also be acquired from the measurement of arterial oxygen saturation. This implies the possibility to acquire the physiological parameters other than arterial oxygen saturation form pulse oximeters.
Download
Paper Nr: 50
Title:

Comparison of Different Classifiers on a reduced set of features
Authors:

Giovanni Saggio, Pietro Cavallo, Lucia R. Quitadamo, Maria G. Marciani, Luigi Bianchi, Gianluca Susi and Giovanni Costantini
Abstract: In this study a comparison among three different machine learning techniques for the classification of mental tasks for a Brain-Computer Interface system is presented: MLP neural network, Fuzzy C-Means Analysis and Support Vector Machine (SVM). In BCI literature, finding the best classifier is a very hard problem to solve, and it is still an open question. We considered only ten electrodes for our analysis, in order to lower the computational workload. Different parameters were analyzed for the evaluation of the performances of the classifiers: accuracy, training time and size of the training dataset. Results demonstrated how the accuracies of the three classifiers are nearly the same but the error margin of SVM on this reduced dataset is larger compared to the other two classifiers. Furthermore neural network needs a reduced number of trials for training purposes, reducing the recording session up to 8 times with respect to SVM and Fuzzy analysis. This suggests how, in the presented case, MLP neural network can be preferable for the classification of mental tasks in Brain Computer Interface systems.
Download
Paper Nr: 52
Title:

Combinatorial Detection of Arrhythmia
Authors:

Spiros Michalakopoulos, Costas S. Iliopoulos, Pascal Ferraro and Julien Allali
Abstract: Three problems that arise from electrocardiogram (ECG) interpretation and analysis are presented, followed by algorithmic solutions based on a combinatorial model. First, the beat classification problem is discussed and possible solutions are investigated. Secondly, given the RR intervals, which can be determined using this combinatorial model, or any QRS detection algorithm, the heart rate is determined in a statistical manner from which sinus bradycardia and sinus tachycardia are inferred. Finally, a new combinatorial method for measuring heart rate variability (HRV) is presented and an algorithm for detecting atrial fibrillation is described. The developed algorithms were implemented and tests were carried out on records from the MIT-BIH arrhythmia database. The results of the tests are presented and discussed.
Download
Paper Nr: 53
Title:

PREREQUISITES FOR AFFECTIVE SIGNAL PROCESSING (ASP) – PART II
Authors:

Egon L. van den Broek, Joris Janssen, Jennifer A. Healey and Marjolein D. van der Zwaag
Abstract: Last year, in van den Broek et al. (2009a), a start was made with defining prerequisites for affective signal processing (ASP). Four prerequisites were identified: validation (e.g., mapping of constructs on signals), triangulation, a physiology-driven approach, and contributions of the signal processing community. In parallel with this paper, in van den Broek et al. (2010) another set of two prerequisites is presented: integration of biosignals and physical characteristic. This paper continues this quest and defines two additional prerequisites: identification of users and theoretical specification. In addition, the second part of a review on the classification of emotions through ASP is presented; the first part can be found in van den Broek et al. (2009a).
Download
Paper Nr: 67
Title:

Synthetic Iris Images from Iris Patterns by means of Evolutionary Strategies
Authors:

Alberto de Santos Sierra, Carmen Sánchez Ávila, Vicente Jara Vera and Javier Guerra Casanova
Abstract: Synthetic Biometric is emerging nowadays as a new research field in biometrics. An artificial iris tissue or a synthetic fingerprint could compromise the security, allowing a non-registered individual to enter the system. However, inverse biometric can also improve current identification systems, enhancing not only its strength against fake-based attacks, but also by replicating unavailable or corrupted data, due to a bad acquisition, for instance. The methods proposed in this document aim to provide a procedure to create a synthetic iris tissue from a stored biometric template, so that a non-registered user could access the system under a registered identity. These algorithms will come out with the result that synthetic sample could be so similar to original as desired.
Download
Paper Nr: 69
Title:

Ratio-Hypothesis-based Fuzzy Fusion with Application to Classification of Cellular Morphologies
Authors:

Xiaobo Zhou and Tuan Pham
Abstract: Fusion of knowledge from multiple sources for pattern recognition has been an active area of research in many scientific disciplines. This paper presents a fuzzy version of a probabilistic fusion scheme, known as permanence-of-ratio-based combination, with application to analysis of cellular imaging for high-content screening. Classification of cellular phenotypes has been carried out to illustrate the usefulness of the permanence-of-ratio-based fuzzy fusion.
Download
Paper Nr: 71
Title:

An Integrated Physiological Model of the Lung Mechanics and Gas Exchange Using Electrical Impedance Tomography in the Analysis of Ventilation Strategies in ARDS Patients

Authors:

A. Wang, Mouloud Denai, Derek Linkens, Mahdi Mahfouf, Ang Wang and Gary Mills
Abstract: Thoracic Electrical Impedance Tomography (EIT) is a non-invasive technique which attempts to reconstruct a cross-sectional image of the internal spatial distribution of conductivity from electrical measurements made by injecting small alternating currents via an electrode array placed on the surface of the thorax. Because air is highly resistive to electric currents whereas fluids and blood are good conductors, it is possible to detect changes in lungs air content with EIT enabling the assessment of ventilation distribution. This paper presents a physiological model which integrates a previously developed gas exchange model with a model of the lung mechanics. This model is combined with a two-dimensional (2D) finite element mesh of the thorax to simulate EIT image reconstruction in patients with acute Respiratory Distress Syndrome (ARDS) under mechanical ventilation. The model was able to track lung ventilation distribution under various simulated ARDS conditions and ventilator settings.
Download
Paper Nr: 72
Title:

TACTILE GUIDANCE OF THE HAND IN A BLIND POINTING TASK: “THE TACTILE COMPASS”
Authors:

Marie-Charlotte Lepelley and Francis Lestienne
Abstract: Using tactile skin receptors that are sensitive to vibrations thereby allowing the use of a “tactile compass” made up of a matrix of micro-vibrators that reproduce tactile encoding on the skin surface to orient the wearer. The tactile compass used in this study consisted in 49 microvibrators laid out in a 7x7 matrix. The 49 microvibrators contained inertial vibrators activated by micromotors. The tactile messages were provided in a dynamic way by the successive activation of each microvibrators. The present study investigated the efficiency of the tactile compass in guiding the hand in a blind pointing task when inserted into an abdominal girdle. More specifically, the performances obtained using tactile coding are compared to those obtained using verbal instructions. The participants had to point, from the central target towards one of the four other targets each corresponding to one of the six directions (upwards, downwards, left, right, backwards and forwards) located either in the frontal plane or in the horizontal plane. Overall, the results reveal the efficiency for gesture guidance of providing tactile messages in a dynamic way, without involving learning. In addition, they establish that tactile information transmitted via our vibrotactile device is involved in the processes of both motor control and production of movement in tridimensional space.
Download
Paper Nr: 74
Title:

SINGLE CHANNEL SOURCE SEPARATION FOR CONVOLUTIVE MIXTURES WITH APPLICATION TO RESPIRATORY SOUNDS
Authors:

A. K. Kattepur, Feng Jin and Farook Sattar
Abstract: In this paper, we attempt to extend single channel source separation techniques to the separation of respiratory sound (RS) and heart sounds (HS). This single channel recording is analyzed and shown to be a convolutive mixture model. After analyzing the reasons for failure of commonly used blind source separation algorithms, we evaluate the efficacy of non-negative matrix factorization (NMF) techniques for this application. Analysis on simulated single channel convolutive mixtures at various sensor positions has been performed. It indicates an average signal to interference ratio (SIR) improvement of greater than 10 dB for the optimal sensor locations. The corresponding range of received power has been also studied for reliable separation of RS and HS. Finally, the proposed model and the NMF separation performance are demostrated to work well on real RS recordings.
Download
Paper Nr: 76
Title:

Matlab Software for Detrended Fluctuation Analysis of Heart Rate Variability
Authors:

João D. Luiz Azevedo Carvalho, Adson Ferreira da Rocha and Fernanda Leite
Abstract: The analysis of heart rate variability (HRV) is an important tool for the assessment of the autonomic regulation of circulatory function. HRV analysis is usually performed using methods that are based on the assumption that the signal is stationary within the experiment duration, which is generally not true for long-duration signals (e.g., 24-hour Holter) or signals acquired during stress tests. This paper presents a Matlab tool for detrended fluctuation analysis (DFA) of HRV signals. DFA is applicable in the context of nonstationary signals, since it involves removing fluctuation trends from the signal. The software is validated using simulated signals with different power-law characteristics, and then demonstrated using real HRV signals, obtained from three groups of subjects: healthy volunteers, individuals with Chagas disease, and individuals with mild to moderate hypertension.
Download
Paper Nr: 79
Title:

Observation with Bounds of Biological Systems: A Linear Programming Approach
Authors:

Fernando Tadeo and Mustapha Ait Rami
Abstract: This paper present a technique to estimate states of biological systems that cannot be directly measured, by using available measurements of other states that affect them. More precisely, the proposed technique makes possible to derive, at each time, the possible range of variation of these unmeasured state. The proposed technique is based on having a compartmental model of the system (that might include uncertainty) and then solving a Linear Programming problem. An application, based on the simulation of the kinetic model of drug distribution through the human body, is given to show the applicability to practical problems.
Download
Paper Nr: 82
Title:

Automatic Screening of Acceleration Signal During Pivot-Shift Test Based on Pearson's Correlation Coefficient
Authors:

Cecilia Signorelli, Nicola Lopomo, Maurilio Marcacci, Simone Bignozzi and Stefano Zaffagnini
Abstract: Anterior cruciate ligament injury produces a pathologic kinematics of the limb that can lead to the evidence of a pivot-shift (PS) phenomenon. PS-test, specifically performed to highlighted this knee dynamic instability, is however difficult to quantify. From a clinical point of view is therefore mandatory to find a set of parameters able to quantitatively characterize PS phenomenon, thus distinguishing between pathologic and healthy knees. This study proposed a methodology able to automatically quantify PS phenomenon, analysing the signal recorded by means of a tri-axial accelerometer while executing PS-test itself. A signal template, which reproduced the 3D acceleration average trend while PS phenomenon occurs, was passed along the signal in order to recognise the presence of similar patterns. The recognition of the signal interesting share was based on the calculation of the Pearson’s correlation coefficient between the template and the corresponding part of the windowed signal. The data acquisition concerning to the first 35 patients was used to testing the template; in this analysis we considered both the data relative to pathologic and healthy knee, as well as pre- and post-anaesthesia data, in order to evaluate the influence of active muscular resistance. The methodology followed had assured a recognition of PS repetitions with an accuracy of 96.7%, a sensitivity of 81.9% and a specificity of 99.3%; therefore can be considered a valid and easily computable method for the automatic screening of the acceleration signal during PS test. In the future this method will be uptake in order to quantify the possibility to discern between pathologic and healthy knee.
Download
Paper Nr: 84
Title:

SVM Evaluation for Brain Computer Interface Systems
Authors:

Daniele Casali, Pietro Cavallo, Maria G. Marciani, Giancarlo Orengo, Giovanni Saggio, Luigi Bianchi, Giovanni Costantini and Mario Salerno
Abstract: A Support Vector Machine (SVM) classification method for data acquired by EEG registration for brain/computer interface systems is here proposed. The aim of this work is to evaluate the SVM performances in the recognition of a human mental task, among others. Such methodology could be very useful in important applications for disabled people. A prerequisite has been the developing of a system capable to recognize and classify the following four tasks: thinking to move the right hand, thinking to move the left hand, performing a simple mathematical operation, and thinking to a nursery rhyme. The data set exploited in the training and testing phases has been acquired by means of 61 EEG electrodes and consists of several time series. These time data sets were then transformed into the frequency domain, in order to obtain the power frequency spectrum. In such a way, for every electrode, 128 frequency channels were obtained. Finally, the SVM algorithm was used and evaluated to get the proposed classification.
Download
Paper Nr: 85
Title:

DISCRIMINATION BETWEEN ISCHEMIC AND HEART-RATE RELATED ST-EPISODES
Authors:

Sebastian Zaunseder, Ruediger Poll and Wolfgang Aipperspach
Abstract: Transient ST-epsiodes recognized in the ECG are regarded as marker of myocardial ischemia. As disturbed ST-sections may appear as ST-episodes a differentiated analysis is necessary to avoid misinterpretations. The presented study aims for the discrimination of ischemic and heart-rate related ST-episodes. Our approach includes the morphologic description of the ventricular repolarization by means of the Karhunen-Loève-Transformation and the non-linear classification using an artificial neural network. The proposed selection of used ECG segments guarantees that the classification procedure indicating ischemic attacks can be done before the complete episode is acquired. This online-capable approach gains accuracies up to 94,2 % for the discrimination of ischemic and heart-rate related ST-episodes.
Download
Paper Nr: 86
Title:

Hybrid Physiological Modeling of Subjects Undergoing Cyclic Physical Loading

Authors:

Mouloud Denai, Mahdi Mahfouf, Ahmed Nassef, Emad Elsamahy and Ching-Hua Ting
Abstract: This paper investigates the influence of physical stress on the physiological parameters of the cardiovascular system (CVS). The work aims at estimating the physiological variables such as the Heart Rate (HR), Blood Pressure (BP), Total Peripheral Resistance (TPR) and respiration for a human being who would be subjected to physical workload. The core of the model was based on the model architecture previously developed by Luczak and his co-workers. Luczak's model was first reconstructed and the original published figure plots were used to identify some of the missing parameters via Genetic Algorithms (GA). The model was then modified using real experimental data extracted from healthy subjects who underwent two-session experiments of cyclic-loading based physical stress. Neuro-Fuzzy models were elicited via the data in order to describe the non-linear components of the model. The model response has also been significantly improved by including a dynamics-based component represented by 'time' as an extra input. The final model, as well as being of a ‘hybrid’ nature, was found to generalize better, to be more amenable to expansions and to also lead to better predictions.
Download
Paper Nr: 88
Title:

Framework of an Estimation Algorithm of Time Varying Multijoint Human Arm Viscoelasticity
Authors:

Mingcong Deng, Akira Yanou and Ni Bu
Abstract: The paper concerns a framework of an estimation of multijoint human arm viscoelasticity in a small sufficient time period. The uncertainties have to be considered in estimating the viscoelasticity of the multijoint human arm. In general, the uncertainties existing in the structure of the human arm and the motor command from the central nervous system are subject to the non-Gaussian noises. A generalized Gaussian ratio function is brought in to deal with the non-Gaussian noises. The momotonicity of the generalized Gaussian ratio function is studied based on the approximation formula of Gamma functions, then a robust condition is proposed for the computation of even moments using shape parameters. That is, we can guarantee the accuracy of the simulation results and experimental results by the robust condition. The effectiveness of the proposed method is confirmed by the experimental results.
Download
Paper Nr: 91
Title:

HERMES:Mobile Balance and Instability Assessment System
Authors:

Hyduke Noshadi, Shaun Ahmadian, Hagop Hagopian, Jonathan Wodbridge, Foad Dabiri, Navid Amini, Majid Sarrafzadeh and Nick Terrafranca
Abstract: In this paper we introduce Hermes, a lightweight smart shoe and its supporting infrastructure aimed at extending instability analysis and human balance monitoring outside of a laboratory environment. By combining embedded sensing, signal processing and modeling techniques we create a scientific tool capable of quantifying high-level measures. The system monitors walking behavior and uses an instability assessment model to generate quantitative value with episodes of activity identified by the physician as important. The model incorporates variability and correlation of features extracted during ambulation that have been identified by geriatric motion study experts as precursors to instability balance abnormality and possible fall risk. Our experiments demonstrate the feasibility of our model and the complimentary role our system can play by providing long-term monitoring of patients outside a hospital setting at a reduced cost, with greater user convenience, and inference capabilities that meet physicians and researchers needs.
Download
Paper Nr: 95
Title:

Wiener-Hammerstein Parameter Estimation using Differential Evolution-Application to Limb Reflex Dynamics
Authors:

Oliver P. Dewhirst, Oliver Dewhirst, David Simpson, Natalia Angarita, Robert Allen and Philip Newland
Abstract: The nonlinear Wiener-Hammerstein model, which consists of a static (no memory) non-linearity sandwiched between two dynamic (with memory) linear elements, provides a parsimonious and accurate model for representing a number of biological systems. In this study we compare the performance of two Wiener-Hammerstein parameter estimation methods; a commonly used nonlinear local optimisation method and the global optimisation method Differential Evolution. The accuracy and convergence properties of the two methods is tested using experimental data collected from the locusts hind limb reflex control system and using computer simulations.
Download
Paper Nr: 105
Title:

A model of primate photoreceptors
Authors:

Hugo R. Gonçalves, Hugo Gonçalves and Miguel Correia
Abstract: As experimental research reveals the biological mechanisms behind the processing done by the retina, complete models of the retina become more and more possible. This paper presents a temporal model of primate photoreceptors inspired by the mechanisms discovered in other species. It implements light adaptation based on pigment bleaching and biochemical reactions. The simulation provides similar results to experiments made in impulse, contrast and sensitivity response curves of primate cones and rods.
Download
Paper Nr: 107
Title:

ACCURATE LOCALIZATION OF CELL NUCLEI IN PAP SMEAR IMAGES USING GRADIENT VECTOR FLOW DEFORMABLE MODELS
Authors:

Christophoros Nikou, Marina Plissiti and Antonia Charchanti
Abstract: In this work, we present an automated method for the detection of cells nuclei boundaries in conventional PAP stained cervical smear images. The proposed method consists of three phases: a) the definition of candidate nuclei centroids set using mathematical morphology, b) the initial approximation of cells nuclei boundaries and c) the application of the Gradient Vector Flow (GVF) snakes for the final estimation of candidate cell nuclei boundaries. It must be noted that the initial approximation of each snake position is obtained automatically, without any observer interference. For the final determination of the nuclei in our images, we perform a fuzzy C-means clustering, using a data set of patterns based on the characteristics of the area enclosed by the final position of the GVF snakes. The proposed method is evaluated using cytological images of conventional PAP smears, which contain 3616 recognized squamous epithelial cells. The results show that the application of the GVF snakes entails in accurate nuclei boundaries, and consequently in the improvement of the performance of the clustering algorithm.
Download
Paper Nr: 108
Title:

SYNTHESIS OF DRIVING SIGNALS FOR MEDICAL IMAGING ANNULAR ARRAYS FROM ULTRASONIC X-WAVE SOLUTIONS
Authors:

Luis Castellanos, Antonio Ramos and Hector Calás
Abstract: An interesting approach for achieving high-resolution in ultrasonic imaging, is producing in real time limited diffraction waves with annular arrays; this potential option was presented by J.-Y. Lu in 1991 as a possible way for collimating ultrasonic beams in medical imaging, approximating classic X waves with a finite aperture 10-annuli array, driven with 0-order X-wave excitation signal generated by solving the isotropic/homogeneous scalar wave equation. However, detailed solutions for a proper array electrical driving in order to form X-wave fields of both pressure and velocity potential have been not still reported. Paper objective is to show a tool to obtain approximated solutions for the inverse problems of synthesizing voltage excitations sets for producing both possible X wave field profiles, and comparatively investigate their distinct beam collimating capacities. All calculations, simulations and analyses were made for an ad-hoc developed 8-rings 2.5 MHz array transducer. Results show field distributions in ultrasonic pressure, created for two driving approaches derived by inverse processing from calculated pressure and velocity potential fields. The good performances resulting in both cases for beam collimation, confirm tool applicability. Results suggest the viability of the proposed procedure as a promising alternative to classical proposals and also for synthesizing special excitations.
Download
Paper Nr: 109
Title:

Mechanomyographic analysis with 0.2 s and 1.0 s time delay after onset of contraction

Authors:

Eddy Krueger, Guilherme Nogueira-Neto, Vera Button, Eduardo Scheeren, Gean Chu and Percy Nohama
Abstract: Muscle contractions generate lateral oscillations and motion artifacts that can be detected by MMG sensors placed in the inner and outer sides of the forearm. These artifacts can significantly affect signal processing and eventually it is necessary to eliminate their influence in order to detect movements reliably. One approach is to respect a time delay after the onset of contraction. This study aimed to evaluate the correlation of 0.2 s and 1.0 s time delays after the onset of contraction during wrist movements. This work respected two different time delays before initiating the signal analysis. Two analysis window lengths were evaluated (0.25 s and 0.50 s). The results showed that there are strong correlations between the acquired signals with both time delays, mainly the devised RZ feature (0.81–0.95). This study was a first approach to determine whether triaxial MMG features can be used for motor prosthesis control. The axial moduli presented strong correlations for all movements and can be productive in future applications.
Download
Paper Nr: 113
Title:

USING BAT AUDITORY MODELLING FOR OBJECT DISCRIMINATION AND ECHO SEPARATION TASKS
Authors:

Dragana Nikolic, Timos Papadopoulos and Robert Allen

Abstract: This paper proposes a computational echolocation model that can be used for discrimination of the specific target object and reconstruction of the acoustic information from multiple overlapping echoes returning back from that target. An acoustic model for estimation of the backscattering impulse responses of a rigid disk has been developed and employed in order to simulate reflection and scattering of FM signal from the disk surface and edges. By rotating the disk around its central axis reflected echo patterns from its edges change allowing for small time variation between arrivals of each individual echo component. This represents the scenario of a flying insect where the distances from the bat to the insect’s head, body and wings are slightly different with each returning a contribution to the overall echo. The echolocation signal obtained from the rotating disk is further encoded into the spectrogram-like format characteristic for the mammalian auditory system. The simulation results of the acoustic and auditory modelling presented in this paper prove that the proposed model is capable to distinguish between overlapping echo components from the spectrogram-like forms of the echolocation signals.
Download
Paper Nr: 114
Title:

Modelling Modified Atmosphere Packaging for Fruits and Vegetables using Membrane Systems
Authors:

Gabi Escuela, Thomas Hinze, Peter Dittrich, Stefan Schuster and Mario Moreno
Abstract: As living materials, post-harvested fruits and vegetables continue their metabolic activity, exhibiting progressive biochemical changes. Optimisation of environmental conditions during storage of these fresh commodities is required in order to increase their shelf life. In this work we use P systems to abstract molecular interactions that occur between plant organ, film and surrounding atmosphere factors involved in fresh fruit and vegetable package designs. The proposed model constitutes a general framework to simulate the dynamical behaviour of these systems, specially due to gas exchanges and temperature fluctuations. Moreover, the model can be extended introducing other variables and processes that affect quality of such produces. This can be considered, to the best of our knowledge, the first contribution of Membrane Computing in Food Engineering.
Download
Paper Nr: 117
Title:

VALIDATION OF AN AUTOMATED SEIZURE DETECTION SYSTEM ON HEALTHY BABIES
Authors:

Andriy Temko, William Marnane, Gordon Lightbody, Geraldine Boylan and Irina Korotchikova
Abstract: Seizures in newborn babies are commonly caused by problems such as lack of oxygen, haemorrhage, meningitis, infection and strokes. The aim of an automated neonatal seizure detection system is to assist clinical staff in a neonatal intensive care unit to interpret the EEG. In this work, the automated neonatal seizure detection system is validated on a set of healthy patients and its performance is compared to the performance obtained on sick patients reported previously. The histogram-based energy normalization technique is designed and applied to EEG signals from healthy patients to cope with montage mismatch. The results on healthy babies compares favourably to those obtained on sick babies. Several useful observations are made which were not possible to obtain by testing on sick babies only such as a practically useful range of probabilistic thresholds, minimum detection duration restriction, and an influence of the database statistics on the system performance.
Download
Paper Nr: 123
Title:

The Effect of Taping on Motion and Plantar Pressure during ankle inversion
Authors:

Jun Akazawa, Jun Akazawa and Ryuhei Okuno
Abstract: In the field of the sports science and the clinical medicine, to prevent the injury of ankle sprain, there have been various taping techniques for fixing ankle joint. The situation in which ankle inversion sprain will be happen easily is a touchdown phase. The effect of ankle taping is fixed the ankle joint, although there is quite a few description about the difference of the motion with the taping. In order to clarify the taping effects, we had examined the characteristic of ankle taping with calculating the distance between the metatarsus first head and the floor using motion analysis systems and measuring planter pressure during the ankle inversion.
Download
Paper Nr: 128
Title:

An assessment procedure involving waveform shapes for pupil light reflex
Authors:

Minoru Nakayama, Wioletta Nowak, Hitoshi Ishikawa and Ken Asakawa
Abstract: The waveforms of Pupillary Light Reflex (PLR) can be analyzed in a diagnostic test that allows for differentiation between disorders affecting photoreceptors and those affecting retinal ganglion cells. This position paper proposes quantitative comparison metrics for waveform shapes using Discrete Fourier Transform (DFT) descriptors (FDs), and another procedure for emphasizing stimuli and subject differences using MultiDimensional Scaling (MDS) and clustering, where dissimilarities between stimuli are defined using descriptors as waveform features. To determine the efficiency of the procedures, a set of PLR data from a conventional experiment for the determination of a melanopsin-associated photoreceptive system was analyzed. Though the captured data was based on single trial for the stimuli, and the number of samples was small, both characteristics of stimuli and subjects were quantitatively extracted using the proposed procedures. Therefore, the possibility of applying the procedures to clinical diagnostics using PLR was examined.
Download
Paper Nr: 131
Title:

CHARACTERIZATION OF MOLECULAR COMMUNICATION CHANNEL FOR NANOSCALE NETWORKS
Authors:

Mohammad Mahfuz, Dimitrios Makrakis and Hussein Mouftah
Abstract: Recently molecular communication is being considered as a new communication physical layer option for nanonetworks. Nanonetworks are based on nanoscale artificial or bio-inspired nanomachines. Traditional communication technologies cannot work on the nanoscale because of the size and power consumption of transceivers and other components. On the other hand, a detailed knowledge of the molecular communication channel is necessary for successful communication. Some recent studies analyzed propagation impairment and its effects on molecular propagation. However, a proper characterization of the molecular propagation channel in nanonetworks is missing in the open literatures. This goes without saying that a molecular propagation channel has to be characterized first before any performance evaluation can be made. Due to the nanoscale dimension of the nanomachines involved in molecular communication a measurement based approach using in vitro experiments is extremely difficult. In addition, a proper tuning of the experimental parameters is mandatory. This is why the authors were motivated to characterize the ‘channel quantum response (CQR)’ or equivalently the ‘throughput response’ of bio-inspired nanonetworks with an alternative approach. This paper considers the molecular channel as particle propagation. The CQR i.e. the throughput response and its characteristics have been found in order to better-understand the molecular channel behavior of nanonetworks.
Download
Paper Nr: 133
Title:

Estimation of Growth of Ovarian Follicles Using Rigid and Elastic Ultrasound Image Registration
Authors:

Sebastijan Šprager, Boris Cigale and Damjan Zazula
Abstract: In this paper, a method for assessment of the ovarian follicle growth is presented. 3D ultrasound volumes of ovaries are processed. Ovarian follicles are shown as hypoechogenic areas in the cross-section images. In first phase, global translations and rotations of two observed follicle constellations from two consecutive ovary examinations are detected. In second phase, detailed local deformations are estimated using elastic registration. The proposed method has been tested using artificial simulated models of ultrasound images of ovaries. Preliminary results shows the proposed method is efficient and reliably detects deformations ovarian follicles cause by their growth.
Download
Paper Nr: 134
Title:

CONCEPTION OF A PHOSPHENE BASED VISUAL SYSTEM: A FUNCTIONNAL APPROACH

Authors:

Guillaume Tatur, Isabelle MARC, Michel Dumas and gerard Dupeyron
Abstract: This work falls within the broader framework of visual prostheses conception for blind people suffering from a retina photoreceptor cells degenerative disease. Studies on the issue of informational content of the prosthetic vision propose, in majority, a simple reduction in the resolution of greyscale images from a single camera. Our work proposes a novel approach based on functional vision. This functional vision is dedicated to specific needs in mobility, and is based on a simple analysis and less ambiguous representation of the scene layout. Emphasis is placed on the usefulness of providing access to 3D information combined with the use of an eye tracking device that should greatly improve spatial knowledge acquisition. This work presents extraction and proper formatting of data from optical sensors in order to establish a coherent vision based on the generation of a limited number of phosphenes.
Download
Paper Nr: 5
Title:

AUTOMATIC IDENTIFICATION OF DNA MARKERS BASED ON FEATURES REDUCTION
Authors:

Jordi Solé-Casals, Carlos M. Travieso-González, Jesús B. Alonso and Miguel A. Ferrer
Abstract: This paper has implemented a feature reduction based on Independent Components Analysis (ICA) and Principal Component Analysis (PCA) for an automatic supervised identification system of Pejibaye palm DNA markers, using an Artificial Neural Network (ANN) as classifier; obtaining 100% for the classes’ identification. The biochemical parameterization proposed, based on 89 RAPD primer markers applied on haplotypes of Pejibaye races, has correctly been proved for its reduction. The computational times have been studied, obtaining results in real time for test mode. Finally the interesting combination of these techniques (biochemical and computational), gives rise to a formulation of an inexpensive and handy method of origin denomination plant certification.
Download
Paper Nr: 15
Title:

WRITER VERIFICATION BASED ON GRAPHOMETRIC FEATURES USING FEED-FORWARD NEURAL NETWORK

Authors:

Carlos F. Romero, Carlos M. Travieso-González, Jesús B. Alonso and Miguel A. Ferrer
Abstract: This paper shows a writer verification automatic system based on a set of graphometric characteristics extracted from handwritten words. That dataset has been tested with our off-line handwritten database, which consists of 110 writers with 10 samples per writer, where a sample is a dataset of 34 words. After our experiments, we have got a verification success rate of 95.63% and Equal Error Rate (EER) of 3.90% is achieved. For previous results, we have used as classifiers a Neural Network, for each writer.
Download
Paper Nr: 20
Title:

Detection of Obstructive Sleep Apnea from the Frequency Analysis of Heart Rate Variability
Authors:

Abraham Otero, Abraham Otero, Francisco J. Coves, Xosé A. Vila and Francisco Palacios
Abstract: This paper presents a new algorithm for the detection of Obstructive Sleep Apnea (OSA) from a single electrocardiogram lead. It is based on the alterations that OSA patients present in the LF and HF bands of the heart rate variability power spectrum. The algorithm calculates the power of the spectrum in two bands that roughly corresponding with the LF and HF bands. Then the ratio between the power of the low band and the power of the high band is obtained; if this ratio is greater than a certain threshold the patient is classified as having OSA, otherwise he/she is classified as not having OSA. The boundaries of the bands and the threshold were obtained by means of a semi-automatic training stage, where the training data set of the Apnea-ECG Database was used. Then the algorithm was validated over the test data set of this database, classifying correctly 29 of 30 recordings.
Download
Paper Nr: 22
Title:

A MORPHING TECHNIQUE TO ESTIMATE LUNG CANCER DEFORMATION DUE TO BREATHING IN RADIOTHERAPIC TREATMENT
Authors:

Sara Ramella, Lucio Trodella, Luigi Battista, Grazia P. Masselli, Sergio Silvestri, Lucio Trodella, Sara Ramella and Luigi Battista
Abstract: A morphing technique aimed to correlate lung cancer patient’s chest cross circumference variations with tumor morphology during quiet respiration is here described. Two CT slices corresponding to the same tumor section are acquired at forced inspiration and forced expiration and correlated with chest circumference values. An image sequence has been obtained by applying a linear morphing transformation. Each image of the sequence has been associated with a chest circumferential value and a sequence subset images corresponding to subject’s tidal volume has then been selected and compared with a CT slice acquired at tidal volume. Images showing the minimum pixel differences with slice at tidal volume were identified and associated with chest circumference values, allowing to estimate in which phase of the breathing period the CT scan was carried out. CT exams in free-breathing and breath-hold conditions have been conducted on a lung cancer patient in order to correlate the acquired slices with the variations of patient’s chest circumference measured with a pneumatic strain gauge. The here described methodology could allow to define the area to be irradiated during a particular phase of the breathing period, considering the cancer area in the morphing simulation frame corresponding to this phase as target.
Download
Paper Nr: 23
Title:

Feature and computational time reduction on hand biometric system
Authors:

Jordi Solé-Casals, Carlos M. Travieso-González, Jesús B. Alonso and Miguel A. Ferrer
Abstract: In real-time biometric systems, computational time is a critical and important parameter. In order to improve it, simpler systems are necessary but without loosing classification rates. In this present work, we explore how to improve the characteristics of a hand biometric system by reducing the computational time. For this task, neural network-multi layer Perceptron (NN-MLP) are used instead of original Hidden Markov Model (HMM) system and classical Principal Component Analysis (PCA) procedure is combined with MLP in order to obtain better results. As showed in the experiments, the new proposed PCA+MLP system achieves same success rate while computational time is reduced from 247 seconds (HMM case) to 7.3 seconds
Download
Paper Nr: 25
Title:

Combination of Covolutive Blind Signal Separation and Wavelet Decomposition to Extract the Atrial Activity in Atrial Fibrillation
Authors:

Carlos Vayá, José Joaquín Rieta and Raul Alcaraz
Abstract: In order to use the ECG as a tool for the characterization of atrial fibrillation (AF), we need to dissociate atrial activity (AA) from ventricular activity. On the other hand, the reduced number of leads recorded from a Holter system limits the necessary spatial diversity required by Blind Source Separation (BSS) techniques to accurately extract the AA. In this work, we propose a new method, the Convolutive Multiband Blind Separation (CMBS), to solve the problem of reduced number of leads by combining the Wavelet transform with the convolutive BSS algorithm Infomax. Our analysis shows up that CMBS improves the extraction performance of AA from Holter systems in comparison with previous extraction methods. This improvement is accomplished in two different scenarios, one for synthetic signals and another one for real signals. A high accuracy of the estimated AA for synthetic and real AF ECG episodes is reached in both scenarios. In addition, results prove that CMBS preserves the original AA spectral parameters.
Download
Paper Nr: 26
Title:

Detection of toothbrushing activity using free-living acceleration data
Authors:

Ruediger Zillmer
Abstract: The present paper discusses the characterisation of toothbrushing activity, using acceleration data collected for 50 subjects in free-living conditions. The data logging is triggered by super-threshold values of acceleration, which can give rise to false activations by non-brushing activities. Due to large intra and inter individual variations, it is not possible to obtain an exhaustive training-set of all activities that trigger the logging. Thus, a structural analysis of appropriate data features is performed, which reveals a clustering of the data. The comparison with brushing activity traces from laboratory experiments allows the identification of toothbrushing activity, while the remainder corresponds to various false activation events like electronic noise or brush handling. The distribution of the resulting toothbrushing activity shows distinct peaks for morning and night brushing activity.
Download
Paper Nr: 34
Title:

AN AUTOMATED APPROACH FOR PREPROCESSING RETINOGRAPHIES
Authors:

Silvia Alayón, José Luis Sánchez, José Sigut, Jorge Marrero and Manuel González
Abstract: A retinography is a retinal photography useful for the precise tracking of any retinal pathology, especially the glaucoma. Although there are sophisticated procedures for studying the evolution of the optic nerve, sometimes it is not feasible the development of a rigorous tracking due to the high number of patients, the high cost of the procedure and the need of high qualified staff. The design of an automated method for detecting this pathology in its first stages through the automated analysis of retinographies could reduce the cost of the process and the number of required specialists. Inspired by this objective, an automated preprocessing method for retinographies is presented in this paper. The proposed methodology combines information of different color spaces for achieving illumination and contrast enhancement.
Download
Paper Nr: 38
Title:

SYNTHESIZED CARDIAC WAVEFORM IN THE EVALUATION OF AUGMENTATION INDEX ALGORITHMS
Authors:

Vânia Almeida, Carlos Correia, Edite Figueiras, Tânia Pereira, Elisabeth Borges, Helena C. Pereira, José B. Simões, João Cardoso and José L. Malaquias
Abstract: We investigate the performance of a new wavelet based algorithm for Augmentation Index (Aix) determination. The evaluation method relies on reference cardiac-like pulses that are synthesized using a weighted combination of exponentially shaped sub-pulses that represent the three main components of real pulses: the systolic stroke, its reflected replica and the carotid reservoir or windkessel effect. The pulses are parameterized so as to reproduce the main types of cardiac waveforms. The values of AIx yielded by the new algorithm are compared with the ones computed directly from the synthesized waveform and with the values produced by standard Probability Density Function (PDF) analysis.
Download
Paper Nr: 39
Title:

Wavelet Sample Entropy Optimization Through Optimal Mother Function Selection for Atrial Fibrillation Analysis
Authors:

Jose Joaquin Rieta and Raul Alcaraz
Abstract: Wavelet Sample Entropy (WSE) has been previously introduced as a successful methodology to predict electrical cardioversion (ECV) outcome of persistent atrial fibrillation (AF). The method estimates AF organization based on the combination of Wavelet decomposition and non-linear regularity metrics, such as Sample Entropy (SampEn). However, WSE has been only computed by applying a specific wavelet function, such as the fourth-order biorthogonal wavelet. In the present work, with the objective of improving WSE robustness and its diagnostic ability in ECV outcome prediction, several orthogonal wavelet families were tested, and their performances were compared. Results indicated that, for all the functions of the same wavelet family, the same sensitivity and specificity were obtained. Additionally, all the wavelet families reached the same diagnostic ability (80.95% sensitivity and 85.71% specificity), being the same patients incorrectly classified by all the families. These results suggest that any wavelet family could be indistinctly used to estimate successfully AF organization with the WSE methodology. As a consequence, the design of a customized wavelet function adapted to the specific characteristics of AA would not improve the WSE diagnostic ability in the prediction of ECV outcome in AF.
Download
Paper Nr: 40
Title:

Assessment of Noise Impact in Sample Entropy for the Non-invasive Organization Estimation of Atrial Fibrillation
Authors:

Jose Joaquin Rieta and Raul Alcaraz
Abstract: In recent studies, Sample Entropy (SampEn) has demonstrated that can be a very promising non-linear index to assess atrial fibrillation (AF) organization from surface ECG recordings. However, non-linear regularity metrics are notably sensitive to noise. Thereby, in the present work, the effect that noise provokes in AF organization estimation based on SampEn is analyzed. Given that AF organization was estimated by computing SampEn over the atrial activity (AA) signal, to evaluate the noise impact on AA regularity, 25 synthetic signals with different organization degrees were generated following a published model. Noise coming from real ECG recordings with different energy levels was added to the synthesized AA signals to obtain different signal to noise ratios (SNR). Results showed that SampEn, i.e., the AA irregularity, increased with noise, thus hiding the differences between organized and disorganized recordings. Precisely, in the presence of noise, SampEn values were increased, in average, by factors of 1.64, 4.46, 9.46 and 14.23 for SNRs of 24, 15, 9 and 3 dB, respectively. As a conclusion, a successful AF organization evaluation via SampEn requires a proper noise reduction in the AA signal.
Download
Paper Nr: 43
Title:

Ultra-wideband Signals for the Detection of Water Accumulations in the Human Body
Authors:

Johannes Schmid, Johannes Schmid, Elena Pancera, Xuyang Li, Lukasz Niestoruk, Stefan Lamparth, Thomas Zwick and Wilhelm Stork
Abstract: In this paper, the concept for an Ultra-wideband (UWB) radar system for the detection and quantification of water accumulations in the human body is presented. With this system, the amount of water in human organs (e.g. the bladder or the lung) can be estimated by processing reflected UWB signals. A simulation-based prove of concept of this approach is presented and it is shown that the system promises a feasible way to implement a mobile on-body water detection system for medical applications. Based on the simulation results, it can be concluded that UWB technology is a very promising opportunity for the realization of a mobile and continuous on-body water detection system that can drastically reduce the costs in different areas in the fields of urology and cardiology.
Download
Paper Nr: 46
Title:

HEART-RATE ADAPTIVE MATCH FILTER BASED PROCEDURE FOR AUTOMATIC DETECTION OF T-WAVE ALTERNANS FROM 24-HOUR ECG RECORDINGS
Authors:

Laura Burattini, Silvia Bini and Roberto Burattini
Abstract: Twenty-four hour T-wave alternans (TWA) analysis is a promising approach for risk stratification, which still remains unpractical because TWA identification algorithms are complex and require long computation time (CT). The aim of the present study was to test the applicability to 24-hour ECG recordings of our heart-rate adaptive match filter (AMF) which allows TWA detection by submitting ECG data to a band-pass filter centered at the TWA fundamental frequency fTWA, equal to a half heart rate. Two implementations are possible: 1) the passing-band is adapted to a varying fTWA value (FA_AMF), and 2) the filter band is fixed while conditioning the ECG data (SA_AMF). Simulated ECG tracings, characterized by no TWA or by different kinds of TWA, and 24-hour ECG recordings from healthy subjects and coronary artery disease patients were used to identify the fastest of these two implementations. Our results yielded the conclusions that the CT of our AMF-based procedure is independent of the amount of TWA present in the tracing, but depends on ECG sample length and filter implementation. If filter-design tools are available while performing ECG analysis, the FA_AMF implementation is to be preferred because its CT is about one third of SA_AMF CT.
Download
Paper Nr: 60
Title:

FC-BASED SEGMENTATION OF JAWTISSUES
Authors:

Roberto Lloréns, Valery Naranjo, Miriam Clemente, Mariano Alcañiz and Salvador Albalat
Abstract: The success of an oral implant surgery is subject to accurate advance planning. For this purpose, it is fundamental that a computer-guided program provides all the available information in a reliable way. Therefore, to plan a suitable implant placement, an accurate segmentation of the tissues of the jaw is necessary. These tissues are the cortical bone, trabecular core and the mandibular canal. The accurate segmentation of the mandibular canal, along which the inferior alveolar nerve crosses the lower arch, is particularly important since an injury to the canal can result in lip numbness. To this date, existing segmentation methods for the jaw requires high human interaction and/or don’t achieve enough accuracy. Our overall aim is to develop an automatic method for the segmentation of the whole jaw, focusing our efforts on achieving very high accuracy and time efficiency. To this end, this paper presents an exhaustive evaluation of fuzzy connectedness object extraction as a plausible segmentation core for this method, basing on the results achieved on 80 CT slices in terms of detection and false alarm probability and merit factor.
Download
Paper Nr: 81
Title:

Fuzzy Hyper-clustering for Pattern Classification in Microarray Gene Expression Data Analysis
Authors:

Jin Liu, Jin Liu and Tuan Pham
Abstract: Based on the motivation by computational challenges in microarray data analysis, we propose a fuzzy hypercluster analysis as a new framework for pattern classification using such type of data. This approach uses hyperplanes to represent the cluster centers in the fuzzy c-means algorithm. We present in this position paper the formulation of a hyperplane-based fuzzy objective function and suggest possible solutions. Fuzzy hyper-clustering approach appears to have potential as a novel alternative to analyze microarray gene expression data. Furthermore, the proposed hyper-clustering algorithm is not only confined to microarray data analysis but can be used as a general approach for classifying closely related features.
Download
Paper Nr: 83
Title:

A MODEL OF TUMOUR SPHEROID RESPONSE TO RADIATION: IDENTIFIABILITY ANALYSIS
Authors:

Federico Papa
Abstract: A spatially uniform model of tumour growth after a single instantaneous radiative treatment is presented in this paper. The ordinary differential equations model presented may be obtained from an equivalent partial derivative equations model, by integration with respect to the radial distance. The main purpose of the paper is to study its identifiability properties. In fact, a preliminary condition, that is necessary to verify before performing the parameter identification, is the global identifiability of a model. A detailed study of the identifiability properties of the model is done pointing out that it is globally identifiable, provided that the responses to two different radiation doses are available.
Download
Paper Nr: 96
Title:

Frequency domain analysis as risk predictor of sudden cardiac death from long-time ECG recordings
Authors:

Raul Alcaraz, Diego García and César Sánchez
Abstract: Sudden Cardiac Death (SCD) is a disease that may not only affect patients with cardiovascular pathologies, but also to apparently healthy patients. Thereby, identification of patients with a high potential of suffering SCD is crucial for their treatment with adequate therapies. To this respect, in the present work, different signal processing tools were applied to surface electrocardiographic (ECG) recordings to develop markers which can clearly differentiate between subjects without cardiovascular pathologies and patients who died of SCD. Precisely, the proposed indexes were the Spectral Concentration (SC) around the main frequency peak, which reached a sensitivity of 100.00% and a specificity of 88.89%, and the Mean Frequency Distance (MFD) between the first spectral peaks, which provided a sensitivity of 95.00% and a specificity of 100.00%.
Download
Paper Nr: 97
Title:

FREQUNCY BANDS EFFECTS ON QRS DETECTION

Authors:

Mohamed Elgendi, Friso De Boer and Mirjam Jonkman
Abstract: In this paper, we investigate the QRS frequency bands in ECG signals. Any QRS detection algorithm accuracy depends on the frequency range of ECG being processed. The QRS complex has different morphology and frequency band for different arrhythmias and noises in ECG signals. A standard bandpass range that maximizes the signal (QRS complex)-to-noise (T-waves, 60 Hz, EMG, etc.) ratio will be useful in ECG monitoring and diagnostic tools. A sensitive QRS detection algorithm has been introduced to compare the performance of using different frequency bands. The results shows that the recommended bandpass frequency range for detecting QRS complexes is 8-20Hz which the best signal-to-noise ratio.
Download
Paper Nr: 101
Title:

PREREQUISITES FOR AFFECTIVE SIGNAL PROCESSING (ASP) – Part III
Authors:

Egon L. van den Broek, Marjolein van der Zwaag and Jennifer A. Healey
Abstract: This is the third part in a series on prerequisites for affective signal processing (ASP). So far, six prerequisites were identified: validation (e.g., mapping of constructs on signals), triangulation, a physiology-driven approach, and contributions of the signal processing community (van den Broek et al., 2009) and identification of users and theoretical specification (van den Broek et al., 2010). Here, two additional prerequisites are identified: integration of biosignals, and physical characteristics.
Download
Paper Nr: 104
Title:

A new approach for epileptic seizure detection using Extreme Learning Machine
Authors:

Yuedong Song, Pietro Lio and Sarita Azad
Abstract: In this paper, we investigate the potential of discrete wavelet transform (DWT), together with a recently-developed machine learning algorithm referred to as Extreme Learning Machine (ELM), to the task of classifying EEG signals and detecting epileptic seizures. EEG signals are decomposed into frequency sub-bands using DWT, and then these sub-bands are passed to an ELM classifier. A comparative study on system performance is conducted between ELM and back-propagation neural networks (BPNN). Results show that the ELM classifier not only achieves better classification accuracy, but also needs much less learning time compared to the BPNN classifier. It is also found that the length of the EEG segment used affects the prediction performance of classifiers.
Download
Paper Nr: 110
Title:

MODEL-MAPPING BASED VOICE CONVERSION SYSTEM (A Novel Approach to Improve Voice Similarity and Naturalness using Model-based Speech Synthesis Techniques)
Authors:

Baojie Li, Dalei Wu and Hui Jiang
Abstract: In this paper we present a novel voice conversion application in which no any knowledge of source speakers is available, but only sufficient utterances from a target speaker and a number of other speakers are in hand. Our approach consists in two separate stages. At the training stage, we estimate a speaker dependent (SD) Gaussian mixture model (GMM) for the target speaker and additionally, we also estimate a speaker independent (SI) GMM by using the data from a number of speakers other than the source speaker. A mapping correlation between the SD and the SI model is maintained during the training process in terms of each phone label. At the conversion stage, we use the SI GMM to recognize each input frame and find the closest Gaussian mixture for it. Next, according to a mapping list, the counterpart Gaussian of the SD GMM is obtained and then used to generate a parameter vector for each frame vector. Finally all the generated vectors are concatenated to synthesize speech of the target speaker. By using the proposed model-mapping approach, we can not only avoid the over-fitting problem by keeping the number of mixtures of the SI GMM to a fixed value, but also simultaneously improve voice quality in terms of similarity and naturalness by increasing the number of mixtures of the SD GMM. Experiments showed the effectiveness of this method.
Download
Paper Nr: 116
Title:

EEG SIGNALS IN EPILEPSY AND MIGRAINE. Analysis and Simulations by Multi-Agent Systems
Authors:

Alessandro Viganò, Neri Accornero and Alfredo Colosimo
Abstract: The preliminary results of some observations carried out on the spectral content of EEG signals from migranious and epileptic individuals and, in particular, on the spatio-temporal correlation of the neuronal activation in the two pathologies, are presented. In the aim to simulate the qualitative features of EEG signals associated to migraine and epilepsy, we used a computational approach based upon Pearson correlations and a Multi Agent System. Our findings, although still not conclusive, revealed considerable heuristic power on the sole assumption of a similar synchronization process of the underlying neuronal population, and may provide in the long term useful hints to a very difficult problem.
Download
Paper Nr: 118
Title:

MODELLING GLYCAEMIA IN ICU PATIENTS
Authors:

Catherine G. Enright, Michael G. Madden, Norm Aleks, Stuart Russell, Geoffrey Manley, John G. Laffey, Brian Harte, Anne Mulvey and Niall Madden
Abstract: Presented in this paper is a Dynamic Bayesian Network approach to predict glycaemia levels in ICU patients. The occurrence of hyperglycaemia is associated with increased morbidity and mortality in critically ill patients. Due to the large inter-patient and intra-patient variability, the sparse nature of observations, inaccuracies in the data and the large number of factors that influence glycaemia, the system being modelled contains several sources of uncertainty. In the context of this uncertainty, the DBN-based system presented here performs extremely well. By using a DBN we integrate multiple strands of temporal evidence, arriving at varying time intervals, to determine the most probable underlying explanations. A key contribution of this work is that it presents a principled technique for recalibration of model parameters from general population-level values to patient-specific values, based entirely on standard real-time measurements from the patient. While in this paper we apply our approach to the glycaemia problem, this approach is equally applicable to other applications where unseen variables must be assessed and individualized in real time.
Download
Paper Nr: 124
Title:

ENSEMBLE APPROACHES TO PARAMETRIC DECISION FUSION FOR BIMODAL EMOTION RECOGNITION
Authors:

Jonghwa Kim and Florian Lingenfelser
Abstract: In this paper, we present a novel multi-ensemble technique for decision fusion of bimodal information. Exploiting the dichotomic property of 2D emotion model, various ensembles are built from given bimodal dataset containing multichannel physiological measures and speech. Through synergistic combination of the ensembles we investigated parametric schemes of decision-level fusion. Up to 18% of improved recognition accuracies are achieved compared to the results from unimodal classification.
Download
Paper Nr: 130
Title:

Towards Computer Diagnosis of Laryngopathies Based on Speech Spectrum Analysis: A Preliminary Approach
Authors:

Krzysztof Pancerz, Jan Warchol and Jaroslaw Szkola
Abstract: The main goal of this paper is to give the outline of a preliminary approach to creating a computer tool being a diagnosis support system for laryngopathies. This approach is based on speech spectrum analysis. A simple parameter based on a statistical approach is calculated. Two diseases are considered: Reinke's edema and laryngeal polyp. The paper presents a medical background, basic problems, a proposed procedure for the computer tool, and experiments carried out using this tool.
Download
Paper Nr: 135
Title:

On Supervised Metrics for Shape Segmentation
Authors:

Miguel Garcia-Silvente and Dibet Garcia
Abstract: Segmentation is one of the most critical steps in image analysis. Also, the quantification of the error commited during this step is not a straightforward task. In this work, the performance of some comparison function or metrics are studied, when just one object appears in the anylized region. We develop a method for rank many validation measures of segmentation algorithms. It is based on thresholding a test image with a range of threshold and to find the middle threshold value when the performance measure is minimum or maximum. The performance is plotted and the first derivate is employed in the ranking construction. We have determined that RDE and MHD are two performance measures that show the best results.
Download