BIOSIGNALS 2013 Abstracts

Full Papers Short Papers
Full Papers
Paper Nr: 4
Title:

Continuous Nasal Airflow Resistance during Birch Pollen Provocation Test
Authors:

Tiina M. Seppänen, Olli-Pekka Alho, Aleksi Laajala, Elina Rahkola and Tapio Seppänen
Abstract: Even 50% of population suffers from allergic symptoms in some countries. There is a need for an objective measurement method giving an accurate, reliable and continuous measurement data about the dynamic nasal function. A novel method to assess unobtrusively the continuous nasal airflow resistance using calibrated respiratory belts is used to produce a continuous nasal airflow resistance during the birch pollen provocation test. Ten birch pollen allergic and eleven non-allergic volunteers were recruited and measured. A statistically significant change in the nasal airflow resistance was found due to the challenge in the allergic group while no statistically significant change was found in the non-allergic group. Unique continuous nasal airflow resistance curves were derived to show the dynamic changes in the nasal airflow resistance during the provocation test. The continuous curves show in great detail fast and slow reactions to nasal provocations, which may be helpful in studying the reactivity of patients. The presented method could increase the reliability and accuracy of diagnostics and assessment of the effect of nasal treatments.
Download
Paper Nr: 12
Title:

Athlete Identification using Acceleration and Electrocardiographic Measurements Recorded with a Wireless Body Sensor
Authors:

Peter Christ, Felix Werner, Ulrich Rückert and Jörg Mielebacher
Abstract: In this paper we propose a biometric method for identifying humans during walking and jogging. We use acceleration and electrocardiographic measurements recorded with a wireless body sensor attached to a chest strap. Our method does not require a particular acquisition setup. Information on the gait style and on the physiology is combined to identify a human despite severe motion related artefacts in the electrocardiograph and variations in the gait patterns. We propose to identify humans using features extracted in time and frequency domain and a standard classifier. With the collected data of 22 subjects on a treadmill at velocities from 3 to 9 km/h we obtained an accuracy of 98.1 %. The sensitivity of the identification ranged between 94.6 to 99.5% for the different subjects and the specificity was higher than 99.7 %.
Download
Paper Nr: 14
Title:

Seven-day Analysis of Atrial Fibrillation and Circadian Rhythms
Authors:

Rebeca Goya-Esteban, Frida Sandberg, Óscar Barquero-Pérez, Arcadio Garcia-Alberola, Leif Sörnmo and José Luis Rojo-Álvarez
Abstract: In the present work, f-wave morphology is characterized by principal component analysis and a novel temporal parameter defined by the cumulative normalized variance of the 3 largest principal components (r3). The 7-day behavior of persistent atrial fibrillation (AF) was studied in 9 patients using r3, AF frequency, and sample entropy (SampEn). Detection of circadian rhythms depended on the parameter considered: rhythms were found in 6 (r3, SampEn) and 5 (AF frequency) patients, but interestingly not always in the same patients. Two patients had significant circadian rhythm in all parameters. When a circadian rhythm was significantly present in 7 days, it was usually only significantly present in some of the 24-h segments. It is concluded that detailed AF characterization can be achieved with complementary parameters.
Download
Paper Nr: 20
Title:

Statistical Modeling of Atrioventricular Nodal Function during Atrial Fibrillation - An Update
Authors:

Valentina D. A. Corino, Frida Sandberg, Federico Lombardi, Luca T. Mainardi and Leif Sörnmo
Abstract: This paper introduces a number of advancements of our recently proposed model of atrioventricular (AV) node function during atrial fibrillation (AF). The model is defined by parameters characterizing the arrival rate of atrial impulses, the probability of an impulse choosing either one of the two AV nodal pathways, the refractory periods of these pathways, and their prolongation. In the updated model, the characterization of AV nodal pathways is made more detailed and the number of pathways is determined by the Bayesian information criterion. The performance is evaluated on ECG data acquired from twenty-five AF patients during rest and head-up tilt test. The results show that the refined AV node model provides significantly better fit than did the original model. During tilt, the AF frequency increased (6:25±0:58 Hz vs. 6:32±0:61 Hz, p < 0:05, rest vs. tilt) and the prolongation of the refractory periods decreased for both pathways (slow pathway: 0:23±0:20 s vs. 0:11±0:10 s, p < 0:001, rest vs. tilt; fast pathway: 0:24±0:31 s vs. 0:16±0:19 s, p < 0:05, rest vs. tilt). These results show that AV node characteristics can be assessed noninvasively for the purpose of quantifying changes induced by autonomic stimulation.
Download
Paper Nr: 23
Title:

Stable Measurement of Blood Flow while Running Using a Micro Blood Flowmeter
Authors:

Wataru Iwasaki, Masaki Nakamura, Takeshi Gotanda, Satoshi Takeuchi, Masutaka Furue, Eiji Higurashi and Renshi Sawada
Abstract: Skin blood flow during exercise has been studied before, with measurements made using laser Doppler blood flowmeters; however, their use was limited to activities with minimal motion, such as riding bicycle ergometers, because conventional devices are large and their measurements easily altered by movements of the optical fiber, rendering them inappropriate for running. We have previously developed a micro integrated laser Doppler blood flowmeter using microelectromechanical systems (MEMS) technology. The micro blood flowmeter is wearable and can measure signal stably even while the wearer is moving. We monitored skin blood flow during running at velocities of 6 km/h, 8 km/h, and 10 km/h, and were successful in measuring a stable signal under these conditions. We found that at the forehead the skin blood flow increases and, in contrast, at the fingertip it initially decreases during running. We also found that the level of these increases and decreases correlated with the running velocity.
Download
Paper Nr: 35
Title:

Single Frequency Approximation of Volume Conductor Models for Deep Brain Stimulation using Equivalent Circuits
Authors:

Christian Schmidt and Ursula van Rienen
Abstract: The objective of this study was to investigate the role of frequency-dependent material properties on the voltage response and neural activation in a volume conductor model for deep brain stimulation (DBS). A finite element model of the brain was developed comprising tissue heterogeneity of gray matter, white matter, and cerebrospinal fluid, which was derived from magnetic resonance images of the SRI24 multi-channel brain atlas. A model of the Medtronic DBS 3387 lead surrounded by an encapsulation layer was positioned in the subthalamic nucleus (STN). The frequency-dependent properties of brain tissue and their single-frequency approximations were modelled as voltage- and current-controlled equivalent circuits. The frequency of best approximation, for which the pulse deviation between the single-frequency and frequency-dependent voltage response were minimal, was computed in a frequency range between 130 Hz and 1:3 MHz. Single-frequency approximations of the DBS pulses and the resulting volume of tissue activated (VTA) were found to be in good agreement with the pulses and VTAs obtained from the frequency-dependent solution. Single-frequency approximations were computed by combining finite element method with equivalent circuits. This method allows a fast computation of the time-dependent voltage response in the proximity of the stimulated target by requiring only one finite element computation.
Download
Paper Nr: 46
Title:

Dynamic Data-based Modelling of Synaptic Plasticity: mGluR-dependent Long-term Depression

Authors:

Tim Tambuyzer, Tariq Ahmed, C. James Taylor, Daniel Berckmans, Detlef Balschun and Jean-Marie Aerts
Abstract: Recent advances have started to uncover the underlying mechanisms of metabotropic glutamate receptor (mGluR) dependent long-term depression (LTD). However, it is not completely clear how these mechanisms are linked and it is believed that several crucial mechanisms still remain to be revealed. In this study, we investigated whether system identification (SI) methods can be used to gain insight into the mechanisms of synaptic plasticity. SI methods have shown to be an objective and powerful approach for describing how sensory neurons encode information about stimuli. However, to the author’s knowledge it is the first time that SI methods are applied to electrophysiological brain slice recordings of synaptic plasticity responses. The results indicate that the SI approach is a valuable tool for reverse engineering of mGluR-LTD responses. It is suggested that such SI methods can aid to unravel the complexities of synaptic function.
Download
Paper Nr: 50
Title:

Feature Extraction and Classification of Biosignals - Emotion Valence Detection from EEG Signals
Authors:

A. M. Tomé, A. R. Hidalgo-Muñoz, M. M. López, A. R. Teixeira, I. M. Santos, A. T. Pereira, M. Vázquez-Marrufo and E. W. Lang
Abstract: In this work a valence recognition system based on electroencephalograms is presented. The performance of the system is evaluated for two settings: single subjects (intra-subject) and between subjects (inter-subject). The feature extraction is based on measures of relative energies computed in short time intervals and certain frequency bands. The feature extraction is performed either on signals averaged over an ensemble of trials or on single-trial response signals. The subsequent classification stage is based on an ensemble classifier, i. e. a random forest of tree classifiers. The classification is performed considering the ensemble average responses of all subjects (inter-subject) or considering the single-trial responses of single subjects (intra-subject). Applying a proper importance measure of the classifier, feature elimination has been used to identify the most relevant features of the decision making.
Download
Paper Nr: 61
Title:

Local PWV and Other Hemodynamic Parameters Assessment - Validation of a New Optical Technique in an Healthy Population
Authors:

T. Pereira, I. Santos, T. Oliveira, P. Vaz, T. Pereira, H. Santos, H. Pereira, V. Almeida, J. Cardoso and C. Correia
Abstract: Presently the interest in non-invasive devices for monitoring the cardiovascular system has increased in importance, especially in the diagnosis of some pathologies. The proposed optical device reveals an attractive instrumental solution for local pulse wave velocity (PWV) assessment and other hemodynamic parameters analysis, such as Augmentation Index (AIx), Subendocardial Viability Ratio (SEVR), Maximum Rate of Pressure Change (dP/dtmax) and Ejection Time Index (ETI). These parameters allow a better knowledge on the cardiovascular condition and management of many disease states. Two studies were performed in order to validate this technology. Firstly, a comparative test between the optical system and a gold-standard in PWV assessment was carried out. Afterwards, a large study was performed in 131 young subjects to establish carotid PWV reference values as well as other hemodynamic parameters and to find correlations between these and the population characteristics. The results allowed the use of this new technique as a reliable method to determine these parameters. For the total of subjects values for carotid PWV vary between 3-7.69 m s-1 a clear correlation with age and smoking status was found out. The Aix varies between -6.15% and 11.46% and exhibit a negative correlation with heart, and dP/dtmax parameter shows a significant decrease with age.
Download
Paper Nr: 73
Title:

Template Ageing in Iris Recognition
Authors:

Adam Czajka
Abstract: The paper presents an iris ageing analysis based on comparison results obtained for three different iris matchers (two of them have not been used earlier in works devoted to iris template ageing). For the purpose of this research we collected an iris ageing database of samples captured even eight years apart. To our best knowledge, this is the only database worldwide of iris images collected with such a large time distance between capture sessions. We evaluated the influence of the intra- vs. inter-session accuracy of the iris recognition, as well as the accuracy between the short term (up to two years) vs. long term comparisons (from 5 to 9 years). The average genuine scores revealed statistically significant differences with respect to the time distance between examined samples (depending on the coding method, we obtained from 3% to 14% of degradation of the average genuine scores). As the highest degradation of matching scores was observed for the most accurate matcher, this may suggest that the iris pattern ages to some extent. This work answers the call for iris ageing-related experiments, presently not numerous due to serious difficulties with collection of sufficiently large databases suitable for ageing research, and limited access to adequate number of iris matchers.
Download
Paper Nr: 87
Title:

Exploratory EEG Analysis using Clustering and Phase-locking Factor
Authors:

Carlos Carreiras, Helena Aidos, Hugo Silva and Ana Fred
Abstract: Emotion recognition is essential for psychological and psychiatric applications and for improving the quality of human-machine interaction. Therefore, a simple and reliable method is needed to automatically assess the emotional state of a subject. This paper presents an application of clustering algorithms to feature spaces obtained from the acquired EEG of subjects performing a stress-inducing task. These features were obtained in three ways: using the EEG directly, using ICA to remove eye movement artifacts, and using EMD to extract data-driven modes present in the signals. From these features, we computed band-power features (BPFs) as well as pairwise phase-locking factors (PLFs), in a total of six different feature spaces. These six feature spaces are used as input to various clustering algorithms. The results of these clustering techniques show interesting phenomena, including prevalence for low numbers of clusters and the fact that clusters tend to be made of consecutive test lines.
Download
Paper Nr: 90
Title:

Array-based Electromyographic Silent Speech Interface
Authors:

Michael Wand, Christopher Schulte, Matthias Janke and Tanja Schultz
Abstract: An electromygraphic (EMG) Silent Speech Interface is a system which recognizes speech by capturing the electric potentials of the human articulatory muscles, thus enabling the user to communicate silently. This study is concerned with introducing an EMG recording system based on multi-channel electrode arrays. We first present our new system and introduce a method to deal with undertraining effects which emerge due to the high dimensionality of our EMG features. Second, we show that Independent Component Analysis improves the classification accuracy of the EMG array-based recognizer by up to 22.9% relative, which is a first example of an EMG signal processing method which is specifically enabled by our new array-based system. We evaluate our system on recordings of audible speech; achieving an optimal average word error rate of 10.9% with a training set of less than 10 minutes on a vocabulary of 108 words.
Download
Short Papers
Paper Nr: 5
Title:

Healthy/Esophageal Speech Classification using Features based on Speech Production and Audition Mechanisms
Authors:

Sofia Ben Jebara
Abstract: This paper focuses on the classification of speech sequences into two classes: healthy speech and esophageal speech. Two kinds of features are selected: those based on speaker speech production mechanism and those using listener auditory system properties. Two classification strategies are used: the Discriminant Analysis and the GMM based bayesian classifier. Experiments, conducted with a large database, show classification accuracy using both features. Moreover, auditory based features are the best since error rates tend to be null.
Download
Paper Nr: 8
Title:

Automatic Burst Detection based on Line Length in the Premature EEG
Authors:

Ninah Koolen, Katrien Jansen, Jan Vervisch, Vladimir Matic, Maarten De Vos, Gunnar Naulaers and Sabine Van Huffel
Abstract: To extract useful information from preterm electroencephalogram (EEG) for diagnosis and long-term prognosis, automated processing of EEG is a crucial step to reduce the workload of neurologists. Important information is contained in the bursts, the interburst-intervals (IBIs) and the evolution of their duration over time. Therefore, an algorithm to automatically detect bursts and IBIs would be of significant value in the Neonatal Intensive Care Unit (NICU). The developed algorithm is based on calculation of the line length to segment EEG into bursts and IBIs. Validating burst detection of this algorithm with expert labelling and existing methods shows the robustness of this algorithm for the patients under test. Moreover, automation is within our grasp as calculated features mimic values obtained by scoring of experts. The outline for successful computer-aided detection of bursting processes is shown, thereby paving the way for improvement of the overall assessment in the NICU.
Download
Paper Nr: 11
Title:

Vocal Fold Stiffness Estimates for Emotion Description in Speech
Authors:

Victoria Rodellar, Daniel Palacios, Elena Bartolomé and Pedro Gómez
Abstract: The present study affords emotional differentiation in speech from the behaviour of the biomechanical stiffness estimates in voice, regarding dispersion and cyclicality. The Glottal Cyclic Parameters are derived from the vibrato correlates found in the Glottal Source reconstructed from the phonated parts of speech and have been shown to be good indices to neurologic disease detection and monitoring. In this paper the application of these parameters to the characterization of the emotional states affecting a speaker when expressing truth opposite to when they believe not saying the truth is explored. The study is based on the reconstruction of the vocal fold stiffness parameters and in the detection of possible deviations induced by emotional tremor and stress from a baseline. The method is validated using results from the analysis of a gender-balanced speaker’s database. Normative values for the different parameters estimated are given and used in contrastive studies of some cases presented to discussion.
Download
Paper Nr: 13
Title:

Wavelet-based Semblance for P300 Single-trial Detection
Authors:

Carolina Saavedra and Laurent Bougrain
Abstract: Electroencephalographic signals are usually contaminated by noise and artifacts making difficult to detect Event-Related Potential (ERP), specially in single trials. Wavelet denoising has been successfully applied to ERP detection, but usually works using channels information independently. This paper presents a new adaptive approach to denoise signals taking into account channels correlation in the wavelet domain. Moreover, we combine phase and amplitude information in the wavelet domain to automatically select a temporal window which increases class separability. Results on a classic Brain-Computer Interface application to spell characters using P300 detection show that our algorithm has a better accuracy with respect to the VisuShrink wavelet technique and XDAWN algorithm among 22 healthy subjects, and a better regularity than XDAWN.
Download
Paper Nr: 26
Title:

Performance Evaluation of Methods for Correcting Ocular Artifacts in Electroencephalographic (EEG) Recordings
Authors:

Murielle Kirkove, Clémentine François, Aurélie Libotte and Jacques G. Verly
Abstract: The presence of ocular artifacts (OA) due to eye movements and eye blinks is a major problem for the analysis of electroencephalographic (EEG) recordings in most applications. A large variety of methods (algorithms) exist for detecting or/and correcting OA's. We identified the most promising methods, implemented them, and compared their performance for correctly detecting the presence of OA's. These methods are based on signal processing “tools” that can be classified into three categories: wavelet transform, adaptive filtering, and blind source separation. We evaluated the methods using EEG signals recorded from three healthy persons subjected to a driving task in a driving simulator. We performed a thorough comparison of the methods in terms of the usual performances measures (sensitivity, specificity, and ROC curves), using our own manual scoring of the recordings as ground truth. Our results show that methods based on adaptive filtering such as LMS and RLS appear to be the best to successfully identify OA's in EEG recordings.
Download
Paper Nr: 31
Title:

A New Local Adaptive Mass Detection Algorithm in Mammograms
Authors:

Ehsan Koozegar, Mohsen Soryani and Ines Domingues
Abstract: Mammography is the most effective procedure for an early detection of breast abnormalities. Masses are a type of abnormality which are very difficult to be visually detected on mammograms. In this paper an efficient method for detection of masses in mammograms is introduced and tested. The algorithm is inspired by binary search and was evaluated both on mini-MIAS and INBreast databases. Mini-MIAS results show that our algorithm outperforms other competing methods. For INBreast database there are no other published mass detection results for comparison, but we believe that our algorithm has good performance.
Download
Paper Nr: 32
Title:

Indices and Repeatability Tests of Cardiovascular Function Performed on the Arterial Distension Waveform - Case Study: Angiography Intervention
Authors:

V. G. Almeida, J. Borba, H. C. Pereira, T. Pereira, J. Cardoso and C. Correia
Abstract: The arterial distension waveform (ADW) analysis is a reliable technique for cardiovascular function assessment. The purpose of this study was to perform the pre-clinical validation of a non-invasive prototype focusing the repeatability tests and cross-relationships between different subject groups. The evaluation focused parameters retrieved from ADW: systolic peak (SP), dicrotic notch (DN), RP (reflection point) and Augmentation index (AI). One hundred and fifty one subjects (61 men and 90 women, aged between 18 and 80 years) were assigned into four groups based on their clinical characteristics. Database is constituted by healthy, hypertensive and subjects that suffer from stenosis. The cross-correlations analysis between groups allows establishing time parameterizations for each one. Furthermore, the differences between the left and right carotid artery suggest intrinsically variability for each one of the subjects. The coefficient of variation (CV) mean value obtained for all measurements was 18.58%, maximum rate of 33.7% and minimum 8.9%. The stenosis case study demonstrate the potentialities of the use of this prototype in the detection of cardiac anomalies by the monitoring of state alterations through RP, SP and DN time parameterizations with visible changes in RP and SP values (after carotid intervention RP appears later than SP, in opposition with values before intervention), while DN associated time changes little. The tests performed on the ADW showed that is possible the reliable measurement of morphological patterns changes.
Download
Paper Nr: 37
Title:

3D Local Binary Pattern for PET Image Classification by SVM - Application to Early Alzheimer Disease Diagnosis
Authors:

Christophe Montagne, Andreas Kodewitz, Vincent Vigneron, Virgile Giraud and Sylvie Lelandais
Abstract: The early diagnostic of Alzheimer disease by non-invasive technique becomes a priority to improve the life of patient and his social environment by an adapted medical follow-up. This is a necessity facing the growing number of affected persons and the cost to our society caused by dementia. Computer based analysis of Fluorodeoxyglucose PET scans might become a possibility to make early diagnosis more efficient. Temporal and parietal lobes are the main location of medical findings. We have clues that in PET images these lobes contain more information about Alzheimer’s disease. We used a texture operator, the Local Binary Pattern, to include prior information about the localization of changes in the human brain. We use a Support Vector machine (SVM) to classify Alzheimer’s disease versus normal control group and to get better classification rates focusing on parietal and temporal lobes.
Download
Paper Nr: 39
Title:

A New Physarum Learner for Network Structure Learning from Biomedical Data
Authors:

T. Schön, M. Stetter, A. M. Tomé and E. W. Lang
Abstract: A novel structure learning algorithm for Bayesian Networks based on a Physarum Learner is presented. The length of the connections within an initially fully connected Physarum-Maze is taken as the inverse Pearson correlation coefficient between the connected nodes. The Physarum Learner then estimates the shortest indirect paths between each pair of nodes. In each iteration, a score of the surviving edges is incremented. Finally, the highest scored connections are combined to form a Bayesian Network. The novel Physarum Learner method is evaluated with different configurations and compared to the LAGD Hill Climber showing comparable performance with respect to quality of training results and increased time efficiency for large data sets.
Download
Paper Nr: 40
Title:

Evaluation of KINECT and SHIMMER Sensors for Detection of Gait Parameters
Authors:

Katja Orlowski and Harald Loose
Abstract: Detecting gait parameters is possible using various sensors based on different physical principles. In our investigation a visual system, the Microsoft KINECT, and an inertial sensor system with SHIMMER 9DoFsensors, are used for capturing the gait of various persons. Both systems have a small form factor and are affordable regarding cost. Hence these are well-suited for mobile applications in the health care environment. Using these low-cost sensor systems, motion capture and analysis can be done in hospitals, physiotherapy units or nursing homes. This paper focusses on the comparison of detected gait parameters by analyzing statistical parameters. The examination of accuracy of both systems is carried out in two steps; first by initially measuring the gait of a small group of volunteers and second of a larger group. The noise is also examined which has to be filtered out in the preprocessing procedure. The choice of filter impacts the detection of gait parameters. As a result the noise is characterized rather nonspecifically in both systems. As expected, the gait parameters determined by the systems are not identical, but similar. The deviations vary in the specific gait parameters; some are less error-prone than other.
Download
Paper Nr: 41
Title:

A Better Understanding of Esophageal Speech Excitation Source Behavior
Authors:

Radhouane Bouazizi and Sofia Ben Jebara
Abstract: Understanding the excitation source of the esophageal speech is a key approach for understanding the esophageal speech. In this paper, we extract the excitation source using an inverse filtering approach and we analyze it. We, for example, show some similarities with an artificial EGG signal. We also detect the closing instants in order to define cycles of opening/closing of esophagus extremity and to recognize the equivalent of glottal cycles. These cycles are classified into different types according to their characteristics. A physical explanation of the esophagus extremity behavior is systematically given at the different steps of the analysis.
Download
Paper Nr: 45
Title:

Fuzzy-enhanced, Real-time Capable Detection of Biological Viruses using a Portable Biosensor
Authors:

Pascal Libuschewski, Dominic Siedhoff, Constantin Timm, Andrej Gelenberg and Frank Weichert
Abstract: This work presents a novel portable biosensor for indirect detection of viruses by optical microscopy. The focus lies on energy-efficient real-time data analysis for automated virus detection. The superiority of our fuzzy-enhanced time-series analysis over hard thresholding is demonstrated. Real-time capability is achieved through general-purpose computing on graphics processing units (GPGPU). It is shown that this virus detec- tion is real-time capable on an off-the-shelf laptop computer, allowing for a wide range of in-field use-cases.
Download
Paper Nr: 55
Title:

Model-based Detection and Analysis of Animal Behaviors using Signals Extracted by Automated Tracking
Authors:

Gennady Denisov, Tomoko Ohyama, Tihana Jovanic and Marta Zlatic
Abstract: Analysis of behaviors of model organisms has a number of applications, particularly to determination of the function of genes and neurons. Drosophila larva is an especially convenient model system for this kind of study because of availability of powerful genetic analysis tools and of automated tracking software that allows high-throughput recording of animal’s shape and position characteristics as time-dependent signals. We have developed an open source software that allows a high-throughput detection and analysis of a comprehensive set of meaningful behaviors of this species. Using the recorded signals as input variables and a set of processing thresholds as parameters, the software employs model-based algorithms to detect the behavioral actions with high accuracy, typically 1-5%. For each detected action it extracts and stores meaningful quantitative features that allow statistical discrimination of mutants from wild type animals and set stage for subsequent application of machine learning techniques to classification of the mutants.
Download
Paper Nr: 58
Title:

Bio-inspired Face Authentication using Multiscale LBP
Authors:

Ayoub Elghanaoui, Nefissa Khiari Hili, Christophe Montagne and Sylvie Lelandais
Abstract: In this paper, we propose a new approach to recognize 2D faces. This approach is based on experiments performed in the field of cognitive science to understand how people recognize a face. To extract features, the image is first decomposed on a base of wavelets using four-level Difference Of Gaussians (DOGs) functions which are a good modeling of human visual system; then different Regions Of Interest (ROIs) are selected on each scale, related to the cognitive method we refer to. After that, Local Binary Patterns (LBP) histograms are computed on each block of the ROIs and concatenated to form the final feature vector. Matching is performed by means of a weighted distance. Weighting coefficients are chosen based on results of psychovisual experiments in which the task assigned to observers was to recognize people. Proposed approach was tested on IV² database and experimental results prove its efficiency when compared to classical face recognition algorithms.
Download
Paper Nr: 60
Title:

Using Path Planning Techniques to Improve Airway Tree Segmentation from CT Images
Authors:

Paolo Cabras and Jan Rosell
Abstract: Virtual Bronchoscopy (VB) permits the preplanning of operations concerning the airways and provides the necessary guidance to reach the pulmonary lesions. Fundamental for a good VB is the reconstruction of a 3D model of the airways from the CT images. Airway segmentation algorithms usually return the biggest detected volume connected to the trachea (the root tree), but many of them also reconstruct during the segmentation process, small parts not connected to the root tree. To overcome this problem this paper proposes a method, based on path planning techniques, that is able to connect the small isolated pieces of bronchi to the terminal points of the root airway tree, taking into account the growing direction of the branches and the gray values of the CT images. As a result, a more complete 3D model of the airways is obtained.
Download
Paper Nr: 62
Title:

Algorithm for Testing Behavioural Phenotypes in a Zebrafish Model of Parkinson’s Disease
Authors:

Angela Pimentel, Hugo Gamboa, Sérgio Reis Cunha and Ana Dulce Correia
Abstract: Parkinson’s disease (PD) is one of the neurodegenerative diseases with an increased prevalence widely studied by the scientific community. Understanding the behaviour related to the disease is an added value for diagnosis and treatment. Thus the use of an animal model for PD that develops similar symptoms to the human being allows to the clinic a larger vision over the health of a patient. Zebrafish can be used to study some human diseases including PD. This work describes the development of an algorithm for the characterization of behaviour in this specie. The biosensor called Marine On-line Biomonitor System (MOBS) is connected electrically to chambers where the specimen of zebrafish moves freely providing a signal that is related with the fish activity. Using the developed algorithm based on signal processing, statistic analysis and machine learning techniques we present classification of a fish as normal or ill and characterize its behaviour.
Download
Paper Nr: 65
Title:

Phase-Rectified Signal Averaging to Evaluate ANS Development in Premature Infants
Authors:

Maristella Lucchini, Devy Widjaja, Carolina Varon, Katrien Jansen and Sabine Van Huffel
Abstract: Aim: Heart Rate Variability (HRV) is determined by the autonomic nervous system (ANS) and a low value of this parameter is related to neurological pathologies and infants mortality. This study aims to assess the utility and the advantages of HRV analysis by means of phase-rectified signal averaging (PRSA), a technique that obtains curves that are useful to determine the development of the ANS in preterm infants, with less obtrusive monitoring compared to electroencephalography. Methods: For a preliminary study, 24-hour ECGs were taken in NICU at the University Hospital in Leuven, from 12 babies: 4 were term, 4 were born preterm but reached a term postmenstrual age, and 4 were preterm. Heart rate tracks of segments of 27 minutes were extracted and analyzed with the PRSA technique. The curves obtained were quantified by the slope and by an acceleration/deceleration related parameter (AC/DC). Two independent analyses on acceleration and deceleration were carried out to visualize the effects of the sympathetic and parasympathetic system separately. Moreover, the immediate response and the response after 5 seconds were taken into account. Results and Conclusion: All the results were compared and validated with traditional HRV parameters. The results of slope and AD/DC in both types of analysis are promising in providing a simple parameter to assess neurological development deficiency in order to allow faster and preventive intervention. Further studies are needed in a larger population.
Download
Paper Nr: 68
Title:

Collective Probabilistic Dynamical Modeling of Sleep Stage Transitions
Authors:

Sergio A. Alvarez and Carolina Ruiz
Abstract: This paper presents a new algorithm for time series dynamical modeling using probabilistic state-transition models, including Markov and semi-Markov chains and their variants with hidden states (HMM and HSMM). This algorithm is evaluated over a mixture of Markov sources, and is applied to the study of human sleep stage dynamics. The proposed technique iteratively groups data instances by dynamical similarity, while simultaneously inducing a state-transition model for each group. This simultaneous clustering and modeling approach reduces model variance by selectively pooling the data available for model induction according to dynamical characteristics. Our algorithm is thus well suited for applications such as sleep stage dynamics in which the number of transition events within each individual data instance is very small. The use of semi-Markov models within the proposed algorithm allows capturing non-exponential state durations that are observed in certain sleep stages. Preliminary results obtained over a dataset of 875 human hypnograms are discussed.
Download
Paper Nr: 69
Title:

ECG Biometrics: Principles and Applications
Authors:

Hugo Silva, André Lourenço, Filipe Canento, Ana Fred and Nuno Raposo
Abstract: Electrocardiographic (ECG) signals have several properties that can greatly complement the existing, and more established biometric modalities. Some of the most prominent properties are the fact that the signals can be continuously acquired using minimally intrusive setups, are not prone to produce latent patterns, and provide intrinsic liveliness detection, opening new opportunities within the area of biometric systems development. The potential impact of this technique extends to a broad variety of application domains, ranging from the entertainment industry, to digital transactions. In this paper, we present a framework for ECG biometrics, with focus on some of the latest developments and future trends in the field, covering multiple aspects of the problem with the aim of a real-world deployment. Our results so far, further reinforce the feasibility and interest of the method in a multibiometrics approach.
Download
Paper Nr: 72
Title:

Snoring Analysis on Full Night Recordings based in the Energy and Entropy in PSG Basal Studies
Authors:

Tiago Marçal, José Basílio Simões, José Moutinho dos Santos, Agostinho Rosa and João Cardoso
Abstract: Snoring is a widely occurring problem in our society and it is highly associated with pathologies like Obstructive Sleep Apnea Syndrome (OSAS) being, usually, one of the first symptoms to appear. Economically, OSAS has a great impact since sleep disorders affect the daily performance of people in their professional activities. The extensive study of snoring evidences may be useful to improve the knowledge of associated pathologies, such as OSAS or others, at an early state. In this work, we study full night sound recordings of patients undergoing polysomnography (PSG) procedures. Recordings are offline processed to characterize time series of snoring events through the record length and correlated with the PSG data. The main goal of the proposed algorithms is to understand the behaviour of the full night sound recording and to identify snoring event patterns that may help and refine the diagnostics process. To achieve this goal, the relationship between the energy and the entropy was studied, for each respiratory event, in both snoring and non-snoring cases. Recordings are offline processed to characterize time series of snoring events through the record length and correlated with the PSG data. In the future, the relationship between these two physical variables can be used to predict the clinical evolution between a simple snorer patient and a patient with OSAS.
Download
Paper Nr: 74
Title:

On Real Time ECG Segmentation Algorithms for Biometric Applications
Authors:

Filipe Canento, André Lourenço, Hugo Silva and Ana Fred
Abstract: Recognizing an individual’s identity through the use of characteristics intrinsic to that subject is a biometric recognition problem with increasingly number of modalities and applications. Recently, the electrical activity of the heart (the Electrocardiogram or ECG) has been explored as an additional modality to recognize individuals. The ECG signal contains several features, which are unique to each individual. The preprocessing of the ECG signal and the feature extraction steps are crucial for biometric recognition to be successful. In fiducial approaches, this last step is accomplished by correctly detecting the heart beats, and performing their segmentation to extract the biometric templates afterwards. In this work, we present an overview of the different steps of an ECG biometric system, focusing on the evaluation and comparison of multiple real-time heart beat detection and ECG segmentation algorithms, and their application to biometric systems. An evaluation and comparison of the algorithms with annotated datasets (MITDB, NSTDB) is presented, and methods to combine them in order to improve performance are discussed.
Download
Paper Nr: 83
Title:

EEG Discrimination with Artificial Neural Networks
Authors:

Sérgio Daniel Rodrigues, João Paulo Teixeira and Pedro Miguel Rodrigues
Abstract: Neurodegenerative disorders associated with aging as Alzheimer’s disease (AD) have been increasing significantly in the last decades. AD affects the cerebral cortex and causes specific changes in brain electrical activity. Therefore, the analysis of signals from the electroencephalogram (EEG) may reveal structural and functional deficiencies typically associated with AD. This study aimed to develop an Artificial Neural Network (ANN) to classify EEG signals between cognitively normal control subjects and patients with probable AD . The results showed that the EEG can be a very useful tool to obtain an accurate diagnosis of AD. The best results were performed using the Power Spectral Density (PSD) determined by Short Time Fourier Transform (STFT) with a ANN developed using Levenberg - Marquardt training algorithm, Logarithmic Sigmoid activation function and 9 nodes in the hidden layer (correlation coefficient training: 0.99964, test: 0.95758 and validation: 0.9653 and with a total of: 0.99245).
Download
Paper Nr: 84
Title:

EMG Onset Detection - Comparison of Different Methods for a Movement Prediction Task based on EMG
Authors:

Marc Tabie and Elsa Andrea Kirchner
Abstract: In this work a study with 8 male subjects was conducted to compare three preprocessing methods for online capable movement prediction based on the recorded electromyogramm (EMG) signals of the right upper limb. One of the compared methods is the widely used Teager Kaiser Energy Operator (TKEO), the other two are a recently proposed method that is based on variance calculation of the signal and the standard deviation. Scope of the work was to show that fast methods, which are required for online processing, have at least the same performance as more classical approaches with higher demands on computational resources like the TKEO. An adaptive threshold was used for onset detection after preprocessing in all compared cases. Comparisons of preprocessing methods were done with respect to the performance in movement prediction and earliness of onset detection. The influence of different movement speeds on the prediction time and the performance were investigated as well. Results presented here show significant differences between the investigated preprocessing methods concerning the prediction time. As a further result of this study it could be shown that different movement speeds also have a significant effect on the prediction time.
Download
Paper Nr: 103
Title:

Comparison of Two Techniques for Lifting Low-lying Objects on a Table - Part II: EMG and Psychological Measurement
Authors:

Angelina Thiers, Harald Loose, Katja Orlowski, Mildred Bläsing and Marco Wallmann
Abstract: The purpose of this study was to determine differences in health benefits and fatigue when using various lifting techniques. Worldwide, back pain is a common disease. In this context, muscular tension in shoulder and neck areas as well as tension-type headaches are the most common side effects. One frequent cause for this pain is connected with the wrong lifting and carrying of loads. To avoid these types of back pain numerous recommendations concerning the right lifting technique already exist. The most common recommendation is that one should use squat lifting instead of stoop lifting. By means of this technique a relief for the back should be obtained. However, these benefits have not been proven yet. For this study eight healthy subjects were evaluated. The test persons had to lift a load for ten minutes. During the lifting task the muscle activity of nine muscles was documented. At the same time, psychological data were collected in a questionnaire. Both, the physiological and psychological data revealed differences between the lifting techniques. During the stoop lifting, a higher burdening of the back muscles was measured. In addition, following the exercise, a greater and prolonged discomfort in the back muscles was documented.
Download
Paper Nr: 110
Title:

Phase-differencing in Stereo Vision - Solving the Localisation Problem
Authors:

J. M. H. du Buf, K. Terzic and J. M. F. Rodrigues
Abstract: Complex Gabor filters with phases in quadrature are often used to model even- and odd-symmetric simple cells in the primary visual cortex. In stereo vision, the phase difference between the responses of the left and right views can be used to construct a disparity or depth map. Various constraints can be applied in order to construct smooth maps, but this leads to very imprecise depth transitions. In this theoretical paper we show, by using lines and edges as image primitives, the origin of the localisation problem. We also argue that disparity should be attributed to lines and edges, rather than trying to construct a 3D surface map in cortical area V1. We derive allowable translation ranges which yield correct disparity estimates, both for left-view centered vision and for cyclopean vision.
Download
Paper Nr: 9
Title:

Voluntary Eye Movement Patterns while Viewing Müller-Lyer Illusion - A New Screening Method for Scotoma Patients
Authors:

Mika Haapala, Antti Rantanen, Aura Falck, Anja Tuulonen, Eero Väyrynen, Tapio Seppänen and Seppo J. Laukka
Abstract: Diabetic maculopathy and especially advanced glaucoma are the most common eye diseases involving scotomas, blind spots in the visual field. The risk of having scotomas increases significantly with age and over 60 million people worldwide suffer from different forms of glaucoma of whom at least are half are not aware of the eye disease. One of the most common scotoma diagnosis tests is perimetry, a visual field test, which produces a complete map of the visual field, but may not be suitable for large screening of population. We are aiming to develop a new portable screening device for cost effective screening of eye diseases. We studied voluntary eye movement patterns evoked by Myller-Lyer illusion figure. Our study material included six scotoma patients (two with Maculopahia Diabetica, and four with Glaucoma simplex) and six control subjects. We recorded eye movement patterns with a portable Tobii T120 eye-tracker system on which a Müller-Lyer figure was projected. As a result, the variation of the y-component of the eye movement trajectory indicates that the scotoma subjects had more vertical variation in their eye movement pattern than the control subjects (P< .01). The preliminary experiment suggests that further prospective studies using our method of analyzing eye movement patterns is warranted with larger sampale sizes and different types and stages of defects.
Download
Paper Nr: 16
Title:

How to Compensate the Effect of using an Incomplete Wavelet Base for Reconstructing an Image? - Application in Psychovisual Experiment
Authors:

Sylvie Lelandais and Justin Plantier
Abstract: One way in psychovisual experiment to understand human visual system is to analyze separately contents of different spatial frequency bands. To prepare images for this purpose, we proceed to a decomposition of the original image by a wavelet transform centered on selected scales. The wavelets used are Difference Of Gaussians (DOG) according to works modeling the human visual system. Before rebuilding the visual stimulus, various transformations can be performed on different scales to measure the efficiency of the observer, for a given task, according to the spatial frequencies used. The problem is that if we use an incomplete wavelet basis during decomposition, there is a significant loss of information between the original image and the reconstructed image. The work presented here offers a way to solve this problem by using coefficients appropriate for each scale during the decomposition step.
Download
Paper Nr: 19
Title:

Differences of Functional Connectivity Brain Network in Emotional Judgment
Authors:

Mehran Amadlou, Kazuko Hiyoshi-Taniguchi, Jordi Solé-Casals, Hironori Fukuyama, Andrzej Cichocki and Francois-Benoît Vialatte
Abstract: Using combined emotional stimuli, combining photos of faces and recording of voices, we investigated the neural dynamics of emotional judgment using scalp EEG recordings. Stimuli could be either combioned in a congruent, or a non-congruent way. As many evidences show the major role of alpha in emotional processing, the alpha band was subjected to be analyzed. Analysis was performed by computing the synchronization of the EEGs and the conditions congruent vs. non-congruent were compared using statistical tools. The obtained results demonstrate that scalp EEG ccould be used as a tool to investigate the neural dynamics of emotional valence and discriminate various emotions (angry, happy and neutral stimuli).
Download
Paper Nr: 22
Title:

Beat Detection Enhancing using AdaBoost
Authors:

Jakub Kuzilek and Lenka Lhotska
Abstract: Beat detection is a basic and fundamental step in electrocardiogram (ECG) processing. In many ECG application time is crucial and slow beat detection algorithm may cause serious problems. Beat detection algorithm desired property is to detect sufficiently large number of QRS complexes with small error in shortest time as possible. Our proposed method tries to combine weak and fast QRS detectors such as amplitude threshold based detector in order to obtain better detection result with very low computational increase. We developed a modified version of the well known AdaBoost algorithm for combining weak QRS detectors. Our algorithm has been compared with the performance of our implementation of the Pan-Tompkins’s beat detection algorithm.
Download
Paper Nr: 24
Title:

A Channel Selection Method for EEG Classification based on Exponentially Damped Sinusoidal Model and Stochastic Relevance Analysis
Authors:

Leonardo Duque Muñoz, Carlos Guerrero-Mosquera and German Castellanos-Dominguez
Abstract: This work introduces a new methodology to select EEG channels related to epileptic seizures by electroencephalogram (EEG) rhythms extraction. Rhythms extraction is an alternative to extract useful information from specific band frequencies, analyze changes in the EEG signals, and detect brain abnormalities. In this approach, the EEG signals are modeled by Exponentially Damped Sinusoidal model (EDS) and the EEG rhythms extraction is based on Stochastic Relevance Analysis (SRA). Achieve results show that EDS model combined with a stochastic relevance measure is a proper alternative for EEG classification of epileptic signals and also could be used for EEG channel selection with seizure activity. The effectiveness of this approach is compared in each experiment with other well known method for feature extraction called as Rhythmic Component Extraction (RCE). This comparison was done based on the performance of the k-NN classifiers and the channels selected were validated by visual inspection and topographic scalp map. The study uses real and multi-channel EEG data and all the experiments have been supervised by an expert neurologist. We conclude that the proposed scheme is a suitable approach for automatic seizure detection at a moderate computational cost, also opening the possibility of formulating new criteria to select, classify or analyze abnormal EEGs channels.
Download
Paper Nr: 27
Title:

Retinal Blood Vessel Segmentation by a MAS Approach
Authors:

Carla Pereira, Jason Mahdjoub, Zahia Guessoum, Luis Gonçalves and Manuel Ferreira
Abstract: Retinal blood vessels segmentation by color fundus images analysis has got huge importance for the diabetic retinopathy early diagnosis. Several interesting computational approaches have been done in this field, but none of them has shown the required performance due to the use of global approaches. Therefore, a new approach is proposed based on an organization of agents enabling vessels detection. This multi-agent approach is preceded by a preprocessing phase in which the fundamental filter is a Kirsch derivative improved version. This first phase allows an environment construction where the agents are situated and interact. Then, blood vessels segmentation emerges from agents’ interaction. According to this study, competitive results were achieved comparing to those found in the present literature. It seems to be that a very efficient system for the diabetic retinopathy diagnosis can be built using MAS mechanisms.
Download
Paper Nr: 30
Title:

Identification of Pronation-supination Patterns on Runners - An Aplication of Functional Principal Component Analysis
Authors:

E. Medina, H. de Rosario, J. Olaso, A. Ballester, J. Navarro and A. Page
Abstract: The correct classification of runners according to their gait patterns is a relevant issue for the design of sports footwear. Specifically, the classification of runners as neutral, pronators, and supinators is a problem that is not yet fully solved, and requires expert observation, since current models based on the automatic processing of kinematic measures are very limited. This work proposes a method based on Functional Data Analysis (FDA) for automatically describing the morphology of the curves that represent ankle movement patterns. By Functional Analysis of Principal Components, the information contained in each data stream is reduced to a small set of variables, that allows an efficient classification of subjects.
Download
Paper Nr: 33
Title:

Chaos and Nonlinear Time-series Analysis of Finger Pulse Waves for Depression Detection
Authors:

Tuan D. Pham, Truong Cong Thang, Mayumi Oyama-Higa, Hoc X. Nguyen, Hameed Saji and Masahide Sugiyama
Abstract: Depressive disorders are mental illnesses that can severely affect one’s health and well-being. If depression is not early detected and left untreated, it can consequently lead to suicide. This paper presents for the first time a novel combination of chaos theory and nonlinear dynamical analysis of signal complexity of photoplethysmography waveforms for detection of depression. Experimental results obtained from the analysis of mentally disordered and control subjects suggest the potential application of the proposed approach.
Download
Paper Nr: 44
Title:

Quantitative Assessment of Diabetics with Various Degrees of Autonomic Neuropathy
Authors:

Chuang-Chien Chiu, Shoou-Jeng Yeh and Yi-Chun Kuo
Abstract: In this study, we investigate the feasibility of using power spectral density (PSD) analysis and cross-correlation function (CCF) analysis to assess the healthy subjects and diabetics with mild, moderate and severe autonomic neuropathy. Continuous cerebral blood flow velocity (CBFV) was measured using transcranial Doppler ultrasound (TCD), and continuous arterial blood pressure (ABP) was measured using Finapres device under supine, tilt-up and hyperventilation conditions. In PSD analysis, the results revealed that the autonomic nervous balance to normal subjects declined in trend from supine to hyperventilation in comparison with that of diabetics. The CCF analysis of mean ABP (MABP) and mean CBFV (MCBFV) for each group of patients was calculated in three frequency bands, i.e., very low frequency (VLF), low frequency (LF), and high frequency (HF). The maximum peak value of CCF (Max CCF) and its corresponding standard deviation and time lag were obtained. Max CCF values at LF of normal subjects and patients with diabetes without autonomic neuropathy in both supine and tilt-up positions were significantly larger than that of diabetics with autonomic neuropathy. Max CCF values gradually increased in hyperventilation at VLF from normal subjects to diabetics without autonomic neuropathy, diabetics with mild autonomic neuropathy, and diabetics with severe autonomic neuropathy.
Download
Paper Nr: 48
Title:

Dominant Frequency, Regularity and Organization Indexes Response to Preprocessing Filter Variations on Simulated Electrograms During Atrial Fibrillation
Authors:

Andrés Orozco-Duque, Juan P. Ugarte, Catalina Tobón, Carlos A. Morillo, Javier Saiz and John Bustamante
Abstract: Atrial fibrillation (AF) is the most common tachyarrhythmia. An accurate AF diagnosis is based on electrograms (EGM) interpretation. In this work, we evaluated different preprocessing filters to calculate dominant frequency (DF), organization index (OI) and regularity index (RI) for EGMduring AF, this comparative analysis has not been reported previously. EGMwere obtained from AF simulated by a three-dimensional model of human atria. The preprocessing stage with a low-pass filter, rectification and high pass filter was implemented. Additionally, a test over 60 simulated signals using different filters with different orders was developed. It was found that some filters affect the DF calculation and cause a bimodal distribution in the DF histogram. Calculation of DF, OI and RI by Fast Fourier Transform analysis of preprocessing signals has high sensibility to filters settings. OI and RI analysis works properly in EGMwith single potentials, however, DF is not necessary related with cycle length in fragmented EGM and OI calculation becomes inaccurate with high DF.
Download
Paper Nr: 49
Title:

Support Vector Machine and Artificial Neural Network Implementation in Embedded Systems for Real Time Arrhythmias Detection
Authors:

Andrés Orozco-Duque, Santiago Rúa, Santiago Zuluaga, Alfredo Redondo, Jose V. Restrepo and John Bustamante
Abstract: This article presents the development and implementation of an artificial neural network (ANN) and a support vector machine (SVM) on a 32-bit ARM Cortex M4 microcontroller core from Freescale Semiconductors and on a FPGA Spartan 6 from Xilinx. The ANN and SVM were implemented for real time detection of ventricular tachycardia (VT) and ventricular fibrillation (VF), and they were compared in terms of accu-racy and computational cost. A Fast Wavelet Transform (FWT) was used, and the energy in each sub-band frequency was calculated in the feature extraction stage. For the training and validation algorithms, signals from MIT-BIH database with normal sinus rhythm, VF and VT in a time window of 2 seconds were used.Test results achieve an accuracy of 99.46% for both ANN and SVM with execution times less than 0.6 ms in microcontroller and 30 µ s in FPGA for ANN and less than 30 ms in a microcontroller for SVM. The test was done with a 32 Mhz clock.
Download
Paper Nr: 54
Title:

EEG Motor Imagery Classification of Upper Limb Movements
Authors:

Maria Claudia F. Castro, João Pedro de O. P. Galhianne and Esther Luna Colombini
Abstract: C EEG channel data are usually used when building systems that aim at distinguishing among right and left hand movements. Few alternatives use multichannel systems when bigger sets of motor imagery are subject to classification and more inputs are required. In this context, this work proposes the use of 8 EEG channels (F,C,P, and O), disposed in a non-conventional set up, to classify up to 4 motor imagery of the upper limbs through a Linear Discriminant Analysis classifier. A spatial feature selection, prior to classification, is applied in order to improve the classification accuracy. For the many channel combinations tested, results suggest that, in addition to the motor areas, other brain areas should be considered. For the proposed system, the best classification accuracy was achieved when distinguishing between left arm and left hand (89.74%) and using only the electrodes in F areas. For the right versus left hand a 71.80% rate was obtained, with electrodes either in P and O areas or in F and P areas. To discriminate between arms and hands, independently of the body side, the best score was 83.33%, for F and P channels, whereas for right and left limbs the best score was 66.02%, with only P channels. The best classification accuracy for the 4 movement problem achieved 50.00%, using all electrodes.
Download
Paper Nr: 63
Title:

A New Approach for the Glottis Segmentation using Snakes
Authors:

G. Andrade Miranda, N. Saenz-Lechón, V. Osma-Ruiz and J. I. Godino-Llorente
Abstract: The present work describes a new methodology for the automatic detection of the glottal space from laryngeal images based on active contour models (snakes). In order to obtain an appropriate image for the use of snakes based techniques, the proposed algorithm combines a pre-processing stage including some traditional techniques (thresholding and median filter) with more sophisticated ones such as anisotropic filtering. The value selected for the thresholding was fixed to the 85% of the maximum peak of the image histogram, and the anisotropic filter permits to distinguish two intensity levels, one corresponding to the background and the other one to the foreground (glottis). The initialization carried out is based on the magnitude obtained using the Gradient Vector Flow field, ensuring an automatic process for the selection of the initial contour. The performance of the algorithm is tested using the Pratt coefficient and compared against a manual segmentation. The results obtained suggest that this method provided results comparable with other techniques such as the proposed in (Osma-Ruiz et al., 2008).
Download
Paper Nr: 66
Title:

A Signal-independent Algorithm for Information Extraction and Signal Annotation of Long-term Records
Authors:

Rodolfo Abreu, Joana Sousa and Hugo Gamboa
Abstract: One of the biggest challenges when analysing data is to extract information from it. In this study, we present a signal-independent algorithm that detects events on biosignals and extracts information from them by applying a new parallel version of the k-means clustering algorithm. Events can be found using a peaks detection algorithm that uses the signal RMS as an adaptive threshold or by morphological analysis through the computation of the signal meanwave. Different types of signals were acquired and annotated by the presented algorithm. By visual inspection, we obtained an accuracy of 97.7% and 97.5% using the L1 and L2 Minkowski distances, respectively, as distance functions and 97.6% using the meanwave distance. The fact that this algorithm can be applied to long-term raw biosignals and without requiring any prior information about them makes it an important contribution in biosignals information extraction and annotation.
Download
Paper Nr: 71
Title:

Detection of Sharp Wave Activity in Biological Signals using Differentiation between Consecutive Samples
Authors:

José L. Ferreira, Pierre J. M. Cluitmans and Ronald M. Aarts
Abstract: A number of signal processing techniques make use of first-derivative-based approaches for detecting regions of interest in biological signals. For instance, central and five-point derivative-based algorithms are employed for emphasizing and identification of the QRS complex in the ECG signal. Signal differentiation approaches are also used for detection and removal of high-frequency components associated to artefacts in the EEG signal. This paper aims to present a first-derivative approach based upon differentiation of consecutive samples – signal slope adaption (SSD) – for detecting regions of sharp wave activity in biological signals. A case study is analysed whereby SSD is used to mark and select the sharp wave activity associated to the QRS complex in the electrocardiogram. Evaluation of our methodology reveals that SSD shows to be effective for identification of QRS samples and, thereby, could be also employed to detect samples associated to sharp wave activity regions of other biological signals which possess similar signal slope behaviour.
Download
Paper Nr: 75
Title:

A Biological Sound Source Localization Model
Authors:

A. Azarfar and J. M. H. du Buf
Abstract: In this paper we address sound source localization in the azimuthal plane. Various models, from the cochlear nuclei to the inferior colliculi, are implemented to achieve accurate and reliable localization. Coincidence detector cells in the medial nuclei and cells sensitive to interaural level difference in the lateral nuclei of the superior olive are combined with models of V- and I-type neurons plus azimuth map cells in the inferior colliculus. An advanced cell distribution in the inferior colliculus is proposed to keep ITD functions at any frequency within the physiological range of the head. Additional projections from the dorsal nucleus of the lateral lemniscus and the medial nucleus of the superior olive are modeled such that interaural time differences in different frequency bands converge to a single result. Experimental results demonstrate good performance in case of a variety of normal sounds.
Download
Paper Nr: 76
Title:

On the Accuracy of Representing Heartbeats with Hermite Basis Functions
Authors:

David G. Márquez, Abraham Otero, Paulo Félix and Constantino A. García
Abstract: Automatic ECG analysis requires choosing a representation for heartbeats. A common approach is using some basis of functions to represent the heartbeat as a linear combination of these functions. The coefficients of the linear combination are used as the features that represent the heartbeat, providing a very compact representation. The most used basis of functions is the one made up of the Hermite functions. Some authors have used as few as 3 Hermite polynomials to represent each heartbeat, while others have used as many as 20. Often little or no justification for the choice of the number of polynomials is given. This paper aims to analyze the impact of using a certain number Hermite polynomials on the accuracy of heartbeat representation. Tests were run fitting the heartbeats of the MIT-BIH arrhythmia database with a number of polynomials ranging from 2 to 20. Three different strategies to determine the heartbeat’s position were used. The fitting errors are reported here. Based on these results, some guidelines to choose a suitable number of Hermite polynomials for different applications are given.
Download
Paper Nr: 77
Title:

A Two-step Subspace Approach for Automatic Detection of CAP Phases in Multichannel Ambulatory Sleep EEG
Authors:

Ammar Hussain Khan, Ibrahim Onaran, Nuri Firat Ince, Mostafa Kaveh, Tacjana Friday, Mike Howell, Thomas Henry and Zhiyi Sha
Abstract: Cyclic Alternating Pattern (CAP) Occurs during Non-Rapid Eye Movement (NREM) Sleep and Is Exploited as a Neuro-Marker of Various Sleep Disorders. the CAP Is Build up from so Called a and B Phases Which Correspond to Widespread Synchronous and Regular Background Activities of EEG Respectively. Currently, These Phases Are Detected by Medical Experts through Visual Inspection, Thereby Limiting Their Potential to Be Used as a Gauge for Sleep Quality. This Paper Aims to Contribute to the Current Effort towards Automatic Detection of CAP Phases, so That Its Potential Can Be Improved in the Assessment of Sleep Quality. unlike Previous Research Where a Predefined Bipolar (and/or Monopolar) Channel Was Used for Automatic Detection, This Paper Explores the Use of a Two-Step Principal Component Analysis (PCA) in Spatial and Feature Domains to Extract Features from All 21 Recording Channels of Ambulatory EEG. Linear Discriminant Analysis (LDA) Was Used on the Extracted Features to Discriminate Phase a and B. over a Five Subject Database, Our Algorithm Reached an Average Classification Accuracy over 86%, Whereas the Baseline Approach Resulted in an 80.3% Success Rate. These Results Indicate That the Two Step PCA Procedure Can Be Used Effectively to Extract Features from Ambulatory EEG towards Detection of CAP.
Download
Paper Nr: 80
Title:

Profiling Arousal in Response to Complex Stimuli using Biosignals
Authors:

Felix Putze, Dominic Heger, Markus Müller, Christian Waldkirch, Yves Chassein, Ivana Kajic and Tanja Schultz
Abstract: We investigate the use of biosignals (blood volume pressure and electrodermal activity) for person-independent profiling of arousal responses to complex, long-term stimuli. We report the design of a user study with 14 subjects to elicitate affective responses with films of different genres. We present a detailed analysis of the recorded signals and show that it is possible extract information on the differences between films and within each film from biosignals. We use this information to automatically discriminate four film classes in a person-independent fashion with an accuracy of up to 97.8%.
Download
Paper Nr: 81
Title:

Eigen Heartbeats for User Identification
Authors:

Marta S. Santos, Ana L. Fred, Hugo Silva and André Lourenço
Abstract: Electrocardiographic (ECG) signals record the heart’s electrical activity over time. These signals have typically been assessed for clinical purposes providing a fair evaluation of the heart’s condition. However, it has been shown recently that they also convey distinctive information that can be used for user identification. In this paper we explore these signals for user identification purposes, proposing two data representation and processing techniques based on principal component analysis (PCA) and classification based on the K-NN rule. We analyze and compare these techniques, showing experimentally that 100% identification rates can be achieved. The analysis covers an outlier removal procedure and different configurations of algorithmic and proposed system parameters.
Download
Paper Nr: 82
Title:

Didactic Speech Synthesizer: Acoustic Module - Formants Model
Authors:

João Paulo Teixeira and Anildo P. Fernandes
Abstract: Text-to-speech synthesis is the main subject treated in this work. It will be presented the constitution of a generic text-to-speech system conversion, explained the functions of the various modules and described the development techniques using the formants model. The development of a didactic formant synthesiser under Matlab environment will also be described. This didactic synthesiser is intended for a didactic understanding of the formant model of speech production.
Download
Paper Nr: 86
Title:

Session-independent EEG-based Workload Recognition
Authors:

Felix Putze, Markus Mülller, Dominic Heger and Tanja Schultz
Abstract: In this paper, we investigate the development of a session-independent EEG-based workload recognition system with minimal calibration time. On a corpus of ten sessions with the same subject, we investigate three different approaches: Accumulation of training data, an adaptive classifier (adaptive LDA) and feature selection algorithm (based on Mutual Information) to improve generalizability of the classifier. In a detailed evalution, we investigate how each approach performs under different conditions and show how we can use those methods to improve classification accuracy by more than 22% and make transfer of models between sessions more reliable.
Download
Paper Nr: 91
Title:

GMM-based Classifiers for the Automatic Detection of Obstructive Sleep Apnea
Authors:

J.-A. Gómez-García, J.-L. Blanco-Murillo, J.-I. Godino-Llorente, L. A. Hernández Gómez and G. Castellanos-Domínguez
Abstract: The aim of automatic pathological voice detection systems is to support a more objective, less invasive diagnosis of diseases. Those detection systems mostly employ an optimized representation of the spectral envelope; whereas for classification, Gaussian Mixture Models are typically used. However, the study of Gaussian Mixture Models-based classifiers as well as Nuisance mitigation techniques, such as those employed in speaker recognition, has not been widely considered in pathology detection tasks. The present work aims at considering whether such tools might improve system performance in detection of pathologies, particularly for the Obstructive Sleep Apnea. Having this in mind, the present paper employs Linear Prediction Coding Coefficients, in conjunction with Gaussian Mixture Model-based classifiers for the detection of Obstructive Sleep Apnea, in a database containing the sustained phonation of vowel /a/. The obtained results demonstrate subtle improvements compared to using baseline automatic detection system.
Download
Paper Nr: 92
Title:

Human Activity Recognition for an Intelligent Knee Orthosis
Authors:

Diliana Rebelo, Christoph Amma, Hugo Gamboa and Tanja Schultz
Abstract: This paper investigates the possibility to classify isolated human activities from biosignal sensors integrated into a knee orthosis. An intelligent orthosis that is capable to recognize its wearers activity would be able to adapt itself to the users situation for enhanced comfort. We use a setup with three modalities: accelerometry, electromyography and goniometry to measure leg motion and muscle activity of the wearer. We segment signals in motion primitives and apply Hidden Markov Models to classify these isolated motion primitives. We discriminate between seven activities like for example walking stairs and ascend or descend a hill. In a small user study we reach an average person-dependent accuracy of 98% and a person-independent accuracy of 79%.
Download
Paper Nr: 95
Title:

Sampling based Optimum Signal Detection in Concentration Encoded Molecular Communication - Receiver Architecture and Performance
Authors:

Mohammad Upal Mahfuz, Dimitrios Makrakis and Hussein T. Mouftah
Abstract: In this paper for the first time ever a comprehensive analysis of the sampling-based optimum signal detection in diffusion-based binary concentration-encoded molecular communication (CEMC) system has been presented. A generalized amplitude shift keying (ASK) based CEMC system has been considered in diffusion-based noise and inter-symbol interference (ISI) conditions. We present an optimum receiver architecture of sampling-based signal detection, address the critical issues in signal detection, and evaluate its performance in terms of sampling number, communication range, and transmission data rate. ISI produced by the residual molecules deteriorates the performance of the CEMC system significantly, which is further deteriorated when the communication distance and/or the transmission data rate increase(s). The proposed receiver architecture can also be used to detect multilevel (M-ary) amplitude modulated signals by increasing the alphabet size and changing the modulation format.
Download
Paper Nr: 97
Title:

Estimation of the Range of Motor Units Firing Rates from EMG Signals using a Fourier-based Power Spectrum Technique
Authors:

Armando Malanda, Ignacio Rodríguez, Luis Gila, Javier Navallas, Javier Rodríguez and Pierre A. Mathieu
Abstract: A method for estimating the range of the motor units mean firing rates from electromyographic (EMG) recordings is presented. The method is based on classical Fourier spectral estimation techniques and is applied to the 0-50 Hz band of the EMG signal within which the mean MU firing rates are usually observed in sustained contractions. Extensive simulations were performed to account for the influence of different signal characteristics such as the firing rate range (FRR), the number of MUAP trains, the coefficient of variation of the motor unit inter-spike intervals (IPI)) and the noise levels. The number of simulated MUAP trains whose mean firing rate dwelled within the estimated range and the estimation error for the lower and upper extremes of the actual FRR were evaluated. While some peaks were undetected and some inaccuracies in the detected firing rate range were observed, satisfactory results were obtained, as for the vast majority of cases the estimated range corresponded to the actual FRR of the simulated MUAP trains.
Download
Paper Nr: 101
Title:

Identification of Molecular Properties Coding Areas in Rat’s Olfactory Bulb by Rank Products
Authors:

Raquel Santano-Martínez, Raquel Leiva-González, Milad Avazbeigi, Agustín González-Gutiérrez and Santiago Marco
Abstract: Neural coding of chemical information is still under strong debate. It is clear that, in vertebrates, neural representation in the olfactory bulb is a key for understanding a putative odour code. To explore this code, in this work we have studied a public dataset of radio images of 2-Deoxyglucose uptake (2-DG) in the olfactory bulb of rats in response to diverse odorants using univariate pixel selection algorithms: rank-products and Mann-Whitney U (MWU) test. Initial results indicate that some chemical properties of odorants preferentially activate certain areas of the rat olfactory bulb. While non-parametric test (MWU) has difficulties to detect these regions, rank-product provides a higher power of detection.
Download
Paper Nr: 102
Title:

Comparison of Two Techniques for Lifting Low-lying Objects on a Table - Part I: Setup, ECG and Motion Measurement
Authors:

Harald Loose, Katja Orlowski, Angelina Thiers and Laura Tetzlaff
Abstract: “There is a strong belief that stoop lifting is ‘bad’ and squat lifting is ‘good’.” In this paper we research a combined motion: lifting and putting a beer crate into a car trunk. This real life task was chosen in the biosignal analysis course at the Brandenburg University of Applied Sciences. We started with the hypothesis that ‘the squat lifting technique is more ergonomic, healthy and less exhausting’. Our study was scheduled for one semester including the experiments and a first preliminary analysis of the data to prove or disprove three partial hypotheses. Four male and four female untrained subjects were involved in the experimental part of the study. Physiological parameters like the heart and the respiration rate, the activity of various muscles as well as the motion of the whole body were measured. Questionnaires were developed and carried out before, immediately after and one week after the experiment to acquire information about the fitness of the subjects and the effects of the exercises on their state of wellness and health. First conclusions result in no clear preference for one lifting technique.
Download
Paper Nr: 104
Title:

Additional Pulmonary Blood Flow in the Cavopulmonary Anastomosis by Means of a Modified Blalock-Taussig Shunt - Is It a Beneficial Clinical Option?
Authors:

Giuseppe D'Avenio, Antonio Amodeo and Mauro Grigioni
Abstract: Since many years, patients with functionally single ventricles are subjected to surgical operations, meant to create a more favourable haemodynamics. The bidirectional cavopulmonary anastomosis (BCPA) is one of such operations, and is principally meant to prepare a future total cavopulmonary anastomosis, i.e., the direct connection of the two vene cavae to the pulmonary arteries. Since the circulation ensuing from a BCPA is basically composed of two circuits in parallel, the upper and the lower circulation, the latter being external to the lung perfusion, there is a potential problem of low oxygen saturation. It has been proposed that an additional pulmonary blood flow, such as that imparted by a modified Blalock-Taussig shunt could be beneficial as for the oxygen saturation. In the present study, this hypothesis is verified by means of a lumped parameter model, considering different degrees of shunting. The results support the view that an additional source of blood flow can have a beneficial effect on the pediatric patient operated on with a BCPA. Future comparison of numerical results with actual clinical data will allow to evaluate the predictive capabilities of the model.
Download
Paper Nr: 105
Title:

Feasibility Study of Heartbeat Detection from Optical Interferometric Signal by using Convolution Kernel Compensation
Authors:

Sebastijan Šprager, Aleš Holobar and Damjan Zazula
Abstract: In this paper, a feasibility of detecting heartbeat from optical interferometric signal by using convolution kernel compensation (CKC) latent variable analysis (LVA) approach is examined. Optical interferometer is a very sensitive device that detects physical elongation of the optical fibre. When used as bed or body sensor, mechanical and audible activity of the heart produce perturbations in the detected signal that, when extracted by LVA, allows completely unobtrusive monitoring of heartbeat. We performed an experiment with fourteen young healthy participants. They exercised on a cycle ergometer until they reached their submaximal heart rate (85 % of maximal heart rate). During resting period after the exercise optical interferometric signal was acquired along with the referential ECG signal. CKC-based decomposition of 1-minute-long signal segments was performed. The obtained efficiency (sensitivity of 97.8 ± 3.0 %, precision of 93.6 ± 7.6 %) and accuracy (reference-to-detected beat delay of 167 ± 65 ms) are within acceptable limits indicating that unobtrusive heartbeat detection using the proposed approach is feasible.
Download
Paper Nr: 107
Title:

Does Low B-value can Handle Q-ball and DTI Reconstructions? - Diffusion MRI Experiment of Ex-vivo Pigs Spinal Cord Phantom
Authors:

Aleksandra Klimas, Kamil Gorczewski, Przemysław Pencak, Zofia Drzazga and Uwe Klose
Abstract: The direction of axons in white matter can be estimated using a deterministic fibre tracking algorithms and diffusion weighted imaging. The aim of this work was to evaluate the data, obtained from pig spines phantom measurements with relatively low b-value, using two types of reconstructions: diffusion tensor imaging (DTI) and q-ball approach. Pigs spines submerged in agar gel were used to prepare a phantom with two crossing populations of fibres. The phantoms were measured in 3T MR scanned for b-value of 1000 and 2000 s/mm2 for q-ball and 200-2000s/mm2 for DTI reconstruction. Analysis of crossing and single fibre population regions in the scanners showed that the median dispersions from the reference directions in case of single fibre population were c.a. 4° and for crossing area c.a. 12° and 6.5° for b-value of 1000 s/mm2 and 2000 s/mm2 respectively. The q-ball approach was able to resolve crossing problem for both low b-values. It was shown here that coherent results can be achieved even with lower b-values than proposed by the theory.
Download