BIOSIGNALS 2017 Abstracts


Full Papers
Paper Nr: 9
Title:

Swarm Intelligence among Humans - The Case of Alcoholics

Authors:

Andrew Schumann and Vadim Fris

Abstract: There are many forms of swarm behaviour, such as swarming of insects, flocking of birds, herding of quadrupeds, and schooling of fish. Sometimes people behave unconsciously and this behaviour of them has the same patterns as behaviour of swarms. For instance, pedestrians behave as herding or flocking, aircraft boarding passengers behave as ant colony, people in escape panic behave as flocking, etc. In this paper we propose a swarm model of people with an addictive behaviour. In particular, we consider small groups of alcohol-dependent people drinking together as swarms with a form of intelligence. In order to formalize this intelligence, we appeal to modal logics K and its modification K'. The logic K is used to formalize preference relation in the case of lateral inhibition in distributing people to drink jointly and the logic K' is used to formalize preference relation in the case of lateral activation in distributing people to drink jointly.

Paper Nr: 10
Title:

Decomposition of the Cardiac and Respiratory Components from Impedance Pneumography Signals

Authors:

Marcel Młyńczak and Gerard Cybulski

Abstract: Impedance pneumography (IP) measures changes of thoracic electrical impedance connected with change of the air volume in the lungs. The electrode configuration used in IP applications causes that electrical heart activity is visible in the IP signals. The aim of this paper is to assess the opportunity to decompose both respiratory and cardiac components and its quality using various methods. Ten students performed static breathing sequences, intended both for calibration and testing. Our prototype, Pneumonitor 2, and the reference pneumotachometer, were used. The accuracy of calculating tidal volume and heart rate, the calibration procedure and the time of analysis, were considered. Mean 86.5% accuracy of tidal volume calculating and only 2.7% error of heart rate estimation were obtained using moving average smoothing filters, for simple short recording of free breathing calibration procedure, in three body positions. More sophisticated adaptive filtering also provided good accuracy, however the processing time was 100-times higher, compared to simple methods. It seems impedance pneumography, without ECG, could be enough for measuring basic cardiorespiratory activity, particularly during ambulatory recordings, in which the least disturbing equipment is desirable.

Paper Nr: 16
Title:

Time-frequency based Coherence and Phase Locking Value Analysis of Human Locomotion Data using Generalized Morse Wavelets

Authors:

Sopapun Suwansawang and David Halliday

Abstract: Time-frequency analysis is a powerful and popular tool for studying time-varying properties of non-stationary neurophysiological signals. In this study, time-frequency based coherence and phase locking value (PLV) analysis using generalized Morse wavelets are presented. The methods are applied to pairs of surface EMG signals recorded from leg muscles during treadmill walking in healthy human subjects. Time-frequency based coherence and PLV analysis in this study detect similar patterns of 8-15 Hz and 15-20 Hz common modulation of EMG during locomotion. Our results suggest that a combination of both methods would be suitable for investigating and characterising non-stationary neurophysiological data. An understanding of the basic principles of normal locomotion can further provide insight into pathological locomotion deficits.

Paper Nr: 17
Title:

Cuff-less Calibration-free Blood Pressure Estimation under Ambulatory Environment using Pulse Wave Velocity and Photoplethysmogram Signals

Authors:

Haruyuki Sanuki, Rui Fukui, Tsukasa Inajima and Shin'ichi Warisawa

Abstract: This paper presents a blood pressure estimation method based on pulse wave velocity (PWV). Although there are a variety of methods based on PWV to estimate blood pressure, most of them require calibration per patient, and the patient has to remain still. The goal of our research is to develop a calibration-free blood pressure estimation method that is applicable not only during rest but also during exercise. To accomplish our goal, we extracted properties of blood vessels from photoplethysmogram (PPG) signals, and compared several regression models, such as the deductive model based on blood vessel physics equation, and the inductive model based on machine learning. Twenty-four participants performed exercise, measuring blood pressure, electrocardiogram (ECG) and PPG. The best result showed that the mean error for the estimated systolic blood pressure (SBP) against cuff-based blood pressure was 0.18 ± 8.68 mmHg. Although there was not a big difference between the regression models, PWV and Augmentation Index are effective features to estimate SBP. In addition to this, Heart Rate was effective only for the young men, and height ratio of c-wave to a-wave of acceleration pulse wave might be effective for elderly men. These results suggest that our proposed method has the potential for cuff-less calibration-free blood pressure estimation which include measurements during rest and exercise.

Paper Nr: 21
Title:

Impact of Airflow Rate on Amplitude and Regional Distribution of Normal Lung Sounds

Authors:

Elmar Messner, Martin Hagmüller, Paul Swatek, Freyja-Maria Smolle-Jüttner and Franz Pernkopf

Abstract: In computerized lung sound research, the usage of a pneumotachograph, defining the phase of respiration and airflow velocity, is essential. To obviate its need, the influence of airflow rate on the characteristics of lung sounds is of great interest. Therefore, we investigate its effect on amplitude and regional distribution of normal lung sounds. We record lung sounds on the posterior chest of four lung-healthy male subjects in supine position with a 16-channel lung sound recording device at different airflow rates. We use acoustic thoracic images to discuss the influence of airflow rate on the regional distribution. At each airflow rate, we observe louder lung sounds over the left hemithorax and a constant regional distribution above an airflow rate of 0.7 l/s. Furthermore, we observe a linear relationship between the airflow rate and the amplitude of lung sounds.

Paper Nr: 23
Title:

A Minimally Invasive Method for Beat-by-Beat Estimation of Cardiac Pressure-Volume Loops

Authors:

Shaun Davidson, Chris Pretty, Shun Kamoi, Thomas Desaive and J. Geoffrey Chase

Abstract: This paper develops a minimally invasive means of estimating a patient-specific cardiac pressure-volume loop beat-to-beat. This method involves estimating the left ventricular pressure and volume waveforms using clinically available information including heart rate and aortic pressure, supported by a baseline echocardiography reading. Validation of the method was performed across an experimental data set spanning 5 Piétrain pigs, 46,318 heartbeats and a diverse clinical protocol. The method was able to accurately locate a pressure-volume loop, identifying the end-diastolic volume, end-systolic volume, mean-diastolic pressure and mean-systolic pressure of the ventricle with reasonable accuracy. While there were larger percentage errors associated with stroke work derived from the estimated pressure-volume loops, there was a strong correlation (average R value of 0.83) between the estimated and measured stroke work values. These results provide support for the potential of the method to track patient condition, in real-time, in a clinical environment. This method has the potential to yield additional information from readily available waveforms to aid in clinical decision making.

Paper Nr: 27
Title:

Characterization of Uterine Response to Misoprostol based on Electrohysterogram

Authors:

C. Benalcazar Parra, R Montfort-Orti, Y. Ye-Lin, J. Alberola-Rubio, A. Perales Marin, J. Mas-Cabo, J. Garcia-Casado and G. Prats-Boluda

Abstract: When the maternal and fetal risks of prolonging gestation are higher than the benefits, labor induction is performed in order to stimulate uterine contractions and to facilitate cervical ripening. Nevertheless, not all cases end up in successful induction leading to an increase in the rate of caesarean sections. The aim of this study was to study the electrophysiological uterine response to misoprostol drug by obtaining and analyzing the evolution of temporal and spectral parameters from uterine electromyogram (electrohysterogram, EHG) records picked up during the first 4 hours of labor induction. Successful inductions showed a progressive increase in amplitude and a frequency shift in spectral content towards higher frequencies approximately 120 min after the initiation of labor induction; such response was not seen in failed inductions. In conclusion, the electrophysiological response caused by effect of misoprostol in pregnant women has been characterized by EHG parameters which showed patterns in their evolution that were different for successful and failed labor inductions. EHG recording and analysis could serve as a very helpful tool to predict the success of labor induction and hence reduce risks and facilitate labor management in this frequent clinical situation.

Paper Nr: 28
Title:

Flow Parameters Derived from Impedance Pneumography after Nonlinear Calibration based on Neural Networks

Authors:

Marcel Młyńczak and Gerard Cybulski

Abstract: Impedance pneumography (IP) is mainly used as a noninvasive method to measure respiratory rate, tidal volume or minute ventilation. It could also register flow-related signals, after differentiation, from spirometrybased forced vital capacity maneuvers or ambulatory-based signals reflecting flow values during natural activity. The aim of this paper is to assess the possibility of improving the accuracy of flow parameters calculated by IP, by using nonlinear neural network correction (as opposed to simple linear calibration), and to evaluate the impact of various calibration procedures and neural network configurations. Ten students carried out fixed static breathing sequences, for both calibration and testing. A reference pneumotachometer and the Pneumonitor 2 were used. The validation of calculating peak and mean flow value during each inspiration and expiration was considered. A mean accuracy of 80% was achieved for a separate neural network with two hidden layers with 10 neurons in each layer, trained individually for each subject and body position, using the data from the longest, fixed calibration procedure. Simple linear modeling achieved only 72.5%.

Paper Nr: 37
Title:

Electrohysterogram Signals from Patients with Threatened Preterm Labor: Concentric Ring Electrode Vs Disk Electrode Recordings

Authors:

Javier Mas-Cabo, Yiyao Ye-Lin, Carlos Benalcazar-Parra, José Alberola-Rubio, Alfredo Perales, Javier Garcia-Casado and Gema Prats-Boluda

Abstract: Recording of electrohysterogram (EHG) has emerged as a non-invasive method for monitoring uterine dynamics during pregnancy. Usually EHG is picked up using conventional disk electrodes placed on the abdominal surface resulting in a limited spatial resolution due to the blurring effect of the volume conductor. In this respect, concentric ring electrodes have been proposed to pick up uterine myoelectrical activity in term patients so as to improve spatial resolution and to reduce physiological interferences embedded in these records. The aim of the present work is to check the feasibility of recording EHG signals using concentric ring electrodes (BC-EHG) in patients with threatened preterm labor and to compare their capability to discriminate true preterm labor from false alarms with that of conventional EHG bipolar recording. For this purpose, 50 sessions with simultaneous EHG recordings with conventional disk electrodes and concentric ring electrodes were conducted in 26 patients. Compared to conventional bipolar EHG recording, the BC-EHG presents smaller amplitude and similar spectral characteristics. Statistically significant differences between women who delivered preterm and those that delivered at term were found for both the average peak-peak amplitude and the dominant frequency in the frequency range 0.2-1 Hz from BC-EHG recordings. Nonetheless no EHG parameter from simultaneous conventional bipolar recording showed statistically significant differences. These results suggest superior performance of BC-EHG recordings in patients with threatened preterm labor for discriminating true preterm labor from term labor.

Paper Nr: 41
Title:

A Neural Network Approach for Automatic Detection of Acoustic Alarms

Authors:

Alex Peiró Lilja, Ganna Raboshchuk and Climent Nadeu

Abstract: Acoustic alarms generated by biomedical equipment are relevant sounds in the noisy Neonatal Intensive Care Unit (NICU) environment both because of their high frequency of occurrence and their possible negative effects on the neurodevelopment of preterm newborns. This work addresses the detection of specific alarms in that difficult environment by using neural network structures. Specifically, both generic and class-specific input models are proposed. The first one does not take advantage of any specific knowledge about alarm classes, while the second one exploits the information about the alarm-specific frequency sub-bands. Two types of partially connected layers were designed to deal with the input information in frequency and in time and reduce the network complexity. The time context was also considered by performing experiments with long short-term memory networks. The database used in this work was acquired in a real-world NICU environment. The reported results show an improvement of more than 9% in absolute value for the generic input model and more than 12% for the class-specific input model, when both consider time information using the proposed partially connected layer.

Paper Nr: 46
Title:

Automated T Wave End Detection Methods - Comparison of Four Different Methods for T Wave End Detection

Authors:

Jonathan Moeyersons, Griet Goovaerts, Suzy Huijghebaert, Bert Vandenberk, Rik Willems and Sabine Van Huffel

Abstract: T wave end detection is essential for electrocardiogram (ECG) processing and analysis. Several methods have been proposed and tested, but an objective comparison is lacking. In this paper, four different (semi-)automated methods are compared with the manually annotated T wave ends of the PhysioNet QT database. The first method is a semi-automatic method, based on a template matching algorithm. The second method uses the tangent of the steepest point of the descending limb of the T wave. The third and fourth method perform a maximum area search of, respectively, a trapezium and the area under the curve. In order to evaluate the accuracy and repeatability of the proposed algorithms, the mean and standard deviation (sd) of the detection errors were computed. This was performed for leads I and II separately, after selection of the best annotated T wave end per beat and after selection of the best lead. We demonstrated that the trapezium method is the least repeatable of all methods tested (sd=29.7ms), whilst the integral method scores best in terms of accuracy (mean=2.2ms). These findings were strengthened by the analysis of the generated Bland-Altman plots, where the smallest bias was observed for the integral method (-1.89ms).

Short Papers
Paper Nr: 6
Title:

DNN-based Models for Speaker Age and Gender Classification

Authors:

Zakariya Qawaqneh, Arafat Abu Mallouh and Buket D. Barkana

Abstract: Automatic speaker age and gender classification is an active research field due to the continuous and rapid development of applications related to humans’ life and health. In this paper, we propose a new method for speaker age and gender classification, which utilizes deep neural networks (DNNs) as feature extractor and classifier. The proposed method creates a model for each speaker. For each test speech utterance, the similarity between the test model and the speaker class models are compared. Two feature sets have been used: Mel-frequency cepstral coefficients (MFCCs) and shifted delta cepstral (SDC) coefficients. The proposed model by using the SDC feature set achieved better classification results than that of MFCCs. The experimental results showed that the proposed SDC speaker model + SDC class model outperformed all the other systems by achieving 57.21% overall classification accuracy.

Paper Nr: 7
Title:

Combining Two Different DNN Architectures for Classifying Speaker’s Age and Gender

Authors:

Arafat Abu Mallouh, Zakariya Qawaqneh and Buket D. Barkana

Abstract: Speakers’ age and gender classification is one of the most challenging problems in the field of speech processing. Recently, remarkable developments have been achieved in the neural network field, nowadays, deep neural network (DNN) is considered one of the state-of-art classifiers which have been successful in many speech applications. Motivated by DNN success, we jointly fine-tune two different DNNs to classify the speaker’s age and gender. The first DNN is trained to classify the speaker gender, while the second DNN is trained to classify the age of the speaker. Then, the two pre-trained DNNs are reused to tune a third DNN (AGender-Tuning) which can classify the age and gender of the speaker together. The results show an improvement in term of accuracy for the proposed work compared with the I-Vector and the GMM-UBM as baseline systems. Also, the performance of the proposed work is compared with other published works on a publicly available database.

Paper Nr: 14
Title:

Wearable Motion Tolerant PPG Sensor for Instant Heart Rate in Daily Activity

Authors:

Takanori Ishikawa, Yasuhide Hyodo, Ken Miyashita, Kazunari Yoshifuji, Yota Komoriya and Yutaka Imai

Abstract: A wristband-type PPG heart rate sensor capable of overcoming motion artifacts in daily activity and detecting heart rate variability has been developed together with a motion artifact cancellation framework. In this work, a motion artifact model in daily life was derived and motion artifacts caused by activity of arm, finger, and wrist were cancelled significantly. Highly reliable instant heart rate detection with high noise-resistance was achieved from noise-reduced pulse signals based on peak-detection and autocorrelation methods. The wristband-type PPG heart rate sensor with our motion artifact cancellation framework was compared with ECG instant heart rate measurement in both laboratory and office environments. In a laboratory environment, mean reliability (percentage of time within 10% error relative to ECG instant heart rate) was 86.5% and the one-day pulse-accuracy achievement rate based on time use data of body motions in daily life was 88.1% or approximately 21 hours. Our device and motion artifact cancellation framework enable continuous heart rate variability monitoring in daily life and could be applied to heart rate variability analysis and emotion recognition.

Paper Nr: 19
Title:

Proposal of New Tracer Concentration Model in Lung PCT Study - Comparison with Commonly Used Gamma-variate Model

Authors:

Maciej Browarczyk, Renata Kalicka and Seweryn Lipiński

Abstract: Perfusion computed tomography (pCT) is one of the methods that enable non-invasive imaging of the hemodynamics of organs and tissues. On the basis of pCT measurements, perfusion parameters such as blood flow (BF), blood volume (BV), mean transit time (MTT) and permeability surface (PS) are calculated and then used for quantitative evaluation of the tissue condition. To calculate perfusion parameters it is necessary to approximate concentration-time curves using regression function. In this paper we compared three regression functions: first commonly used gamma-variate function, second and third Gauss and Rayleigh functions, not previously used for this purpose. The Gauss function showed clear advantage over the others when considering results of simulated data analysis. Actual measurements analysis confirmed conclusions from simulated data analysis. It was showed that contrary to widely accepted belief, the differences between rising and falling edge slope angles of concentration-time curves are inconsiderable. For that reason, it can be assumed that rising and falling edges are symmetrical. The main conclusion is that the Gauss function gives a more robust fit than the widely used gamma-variate function when modelling concentration-time curves in lung pCT studies.

Paper Nr: 20
Title:

Towards Long-term Monitoring of Atrial Fibrillation using Photoplethysmography

Authors:

Birutė Paliakaitė, Andrius Petrėnas, Jurgita Skibarkienė, Tomas Mickus, Saulius Daukantas, Raimondas Kubilius and Vaidotas Marozas

Abstract: This study investigates the feasibility of long-term monitoring of atrial fibrillation (AF) using wrist-worn device, capable of acquiring photoplethysmogram (PPG) and motion data. Moreover, the performance of AF detectors, initially developed to detect AF in electrocardiogram (ECG) signals, is evaluated on PPG. The study population consisted of 12 patients undergoing cardiac rehabilitation. Based on accelerometer data, 65% of recording time was considered as motion-free, which resulted in 86.8 hours of data with AF and 85.4 hours without. The performance of AF detectors was found to be comparable when both ECG and PPG are used for constructing heart rhythm series. Considering that 2/3 of monitoring time PPG was of satisfactory quality, the wrist-worn device has potential to be applied for long-term mass screening of target population.

Paper Nr: 32
Title:

An Anthropomorphic Perspective for Audiovisual Speech Synthesis

Authors:

Samuel Silva and António Teixeira

Abstract: In speech communication, both the auditory and visual streams play an important role, ensuring both a certain level of redundancy (e.g., lip movement) and transmission of complementary information (e.g., to emphasize a word). The common current approach to audiovisual speech synthesis, generally based on data-driven methods, yields good results, but relies on models controlled by parameters that do not relate with how humans do it, being hard to interpret and adding little to our understanding of the human speech production apparatus. Modelling the actual system, adopting an anthropomorphic perspective would provide a myriad of novel research paths. This article proposes a conceptual framework to support research and development of an articulatory-based audiovisual speech synthesis system. The core idea is that the speech production system is modelled to produce articulatory parameters with anthropomorphic meaning (e.g., lip opening) driving the synthesis of both the auditory and visual streams. A first instantiation of the framework for European Portuguese illustrates its viability and constitutes an important tool for research in speech production and the deployment of audiovisual speech synthesis in multimodal interaction scenarios, of the utmost relevance for the current and future complex services and applications.

Paper Nr: 39
Title:

Actigraphic Sleep Detection for Real-World Data of Healthy Young Adults and People with Alzheimer’s Disease

Authors:

Stefan Lüdtke, Albert Hein, Frank Krüger, Sebastian Bader and Thomas Kirste

Abstract: Actigraphy can be used to examine the sleep pattern of patients during the course of the day in their common environment. However, conventional sleep detection algorithms may not be appropriate for real-world daytime sleep detection, since they tend to overestimate the sleep duration and have only been validated for nighttime sleep in a laboratory setting. Therefore, we evaluated the performance of a set of new sleep detection algorithms based on machine learning methods in a real-world setting and compared them to two conventional sleep detection algorithms (Cole’s algorithm and Sadeh’s algorithm). For that, we performed two studies with (1) healthy young adults and (2) nursing home residents with Alzheimer’s dementia. The conventional algorithms performed poorly for these real-world data sets, because they are imbalanced with respect to sensitivity and specificity. A more balanced Hidden Markov Model-based algorithm surpassed the conventional algorithms for both data sets. Using this algorithm leads to an improved accuracy of 4.1 percent points (pp) and 23.5 pp, respectively, compared to the conventional algorithms. The Youden-Index improved by 7.3 and 7.7, respectively. Overall, for a real-world setting, the HMM-based algorithm achieved a performance similar to conventional algorithms in a laboratory environment.

Paper Nr: 42
Title:

Determination of Directional Influences of Kinematic Data in the Stance Period During Running

Authors:

Giovana Yuko Nakashima, Theresa Helissa Nakagawa, Ana Flávia dos Santos, Fábio Viadanna Serrão and Carlos Dias Maciel

Abstract: The study of the interactions among elements of a system is decisive to understanding their behavior patterns. The knowledge of the details of human motion allows physiotherapists to propose prevention and rehabilitation programs, as well as to identify movements that could lead to an injury. This work examines Partial Directed Coherence measures to determine the direction of the influences, throughout the stance phase only, among kinematic joints data acquired during the running activity. Five channels of the ankle, knee, hip, pelvis and trunk kinematic data were processed in each of the three anatomical planes, sagittal, frontal and transverse. These analysis suggested that the ankle joint receives a intense proximal to distal influence, whereas the knee, hip, pelvis and trunk joints presents a predominance of distal to proximal interaction.

Paper Nr: 52
Title:

Comparing Machine Learning Approaches for Fall Risk Assessment

Authors:

Joana Silva, João Madureira, Cláudia Tonelo, Daniela Baltazar, Catarina Silva, Anabela Martins, Carlos Alcobia and Inês Sousa

Abstract: Traditional fall risk assessment tests are based on timing certain physical tasks, such as the timed up and go test, counting the number of repetitions in a certain time-frame, as the 30-second sit-to-stand or observation such as the 4-stage balance test. A systematic comparison of multifactorial assessment tools and their instrumentation for fall risk classification based on machine learning approaches were studied for a population of 296 community-dwelling older persons aged above 50 years old. Using features from inertial sensors and a pressure platform by opposition to using solely the tests scores and personal metrics increased the F-Score of Naïve Bayes classifier from 72.85% to 92.61%. Functional abilities revealed higher association with fall level than personal conditions such as gender, age and health conditions.

Paper Nr: 55
Title:

Heuristic Approximation of the MAP Estimator for Automatic Two-channel Sleep Staging

Authors:

Shirin Riazy, Tilo Wendler, Jürgen Pilz, M. Glos and T. Penzel

Abstract: In this paper, we shall introduce an algorithm that classifies EEG data into five sleep stages, relying only on two-channel sleep measurements. The sleep of a patient (divided into intervals of 30 seconds) is assumed to be a Markov chain on the five-element state space of sleep stages and our aim is to compute the most probable chain of this hidden Markov model by a maximum a posteriori (MAP) estimation in the Bayesian framework. Both the prior distribution of the chains and the likelihood model have to be trained on manual classifications made by professionals. For this purpose, the data is first preprocessed by a Fourier transform, a log transform and a principal component analysis for dimensionality reduction. Since the number of possible chains is immense (roughly 10^335), a heuristic approach for the computation of the MAP estimator is introduced, that systematically discards unlikely chains. The sleep stage classification is then compared to the classification of a professional, who scores according to the AASM and uses a full polysomnography. The overall structure of the hypnogram can adequately be reconstructed with error rates around 25%.

Paper Nr: 56
Title:

Animal Sound Classification using Sequential Classifiers

Authors:

Javier Romero, Amalia Luque and Alejandro Carrasco

Abstract: Several authors have shown that the sounds of anurans can be used as an indicator of climate change. But the recording, storage and further processing of a huge number of anuran's sounds, distributed in time and space, are required to obtain this indicator. It is therefore highly desirable to have algorithms and tools for the automatic classification of the different classes of sounds. In this paper five different classification methods are proposed, all of them based on the data mining domain, which try to take advantage of the sound sequential behaviour. Its definition and comparison is undertaken using several approaches. The sequential classifiers have revealed that they can obtain a better performance than their non-sequential counterpart. The sliding window with an underlying decision tree has reached the best results in our tests, even overwhelming the Hidden Markov Models usually employed in similar applications. A quite remarkable overall classification performance has been obtained, a result even more relevant considering the low quality of the analysed sounds.

Paper Nr: 58
Title:

Iris Recognition under Biologically Troublesome Conditions - Effects of Aging, Diseases and Post-mortem Changes

Authors:

Mateusz Trokielewicz, Adam Czajka and Piotr Maciejewicz

Abstract: This paper presents the most comprehensive analysis of iris recognition reliability in the occurrence of various biological processes happening naturally and pathologically in the human body, including aging, illnesses, and post-mortem changes to date. Insightful conclusions are offered in relation to all three of these aspects. Extensive regression analysis of the template aging phenomenon shows that differences in pupil dilation, combined with certain quality factors of the sample image and the progression of time itself can significantly degrade recognition accuracy. Impactful effects can also be observed when iris recognition is employed with eyes affected by certain eye pathologies or (even more) with eyes of the deceased subjects. Notably, appropriate databases are delivered to the biometric community to stimulate further research in these utterly important areas of iris biometrics studies. Finally, some open questions are stated to inspire further discussions and research on these important topics. To Authors’ best knowledge, this is the only scientific study of iris recognition reliability of such a broad scope and novelty.

Posters
Paper Nr: 3
Title:

Linking Non-Extensive Entropy with Lempel-ziv Complexity to Obtain the Entropic Q-index from EEG Signals

Authors:

Ernane José Xavier Costa, Adriano Rogeri Bruno Tech and Ana Carolina Sousa Silva

Abstract: Physiological data is generated by process that are either nonlinear deterministic or nondeterministic. The lempel-ziv complexity and non-extensive entropy measurement has been used to quantify information in physiological data like EEG and EMG. When the functions of brain cells are affected by damage caused by several disease it is observed changes in the features of the EEG providing useful insight into brain functions and playing a useful role as a first line of decision-support tool for early detection and diagnosis in brain diseases. This paper uses a method to identify the q-index in those signals by using the relationships between entropy definitions given by Lempel-ziv and those given by Tsallis methods. After all, this article shows that, the q-index can be used to characterize EEG seizure quantifying changes related to the q-entropic index.

Paper Nr: 8
Title:

Characterisation of Resting Brain Network Topologies across the Human Lifespan with Magnetoencephalogram Recordings: A Phase Slope Index and Granger Causality Comparison Study

Authors:

Elizabeth Shumbayawonda, Alberto Fernández, Javier Escudero, Michael Pycraft Hughes and Daniel Abásolo

Abstract: This study focuses on the resting state network analysis of the brain, as well as how these networks change both in topology and location throughout life. The magnetoencephalogram (MEG) background activity from 220 healthy volunteers (age 7-84 years), was analysed combining complex network analysis principles of graph theory with both linear and non-linear methods to evaluate the changes in the brain. Granger Causality (GC) (linear method) and Phase Slope Index (PSI) (non-linear method) were used to observe the connectivity in the brain during rest, and as a function of age by analysing the degree, clustering coefficient, efficiency, betweenness, modularity and maximised modularity of the observed complex brain networks. Our results showed that GC showed little linear causal activity in the brain at rest, with small world topology, while PSI showed little information flow in the brain, with random network topology. However, both analyses produced complementary results pertaining to the resting state of the brain.

Paper Nr: 24
Title:

A Low-cost Life Sign Detection Method based on Time Series Analysis of Facial Feature Points

Authors:

Timon Bloecher, Leyre Garralda Iriarte, Johannes Schneider, Christoph Zimmermann and Wilhelm Stork

Abstract: The use of image based presentation attack detection (PAD) systems has experienced an enormous growth of interest in recent years. The most accurate techniques in literature addressing this topic rely on the verification of the actual three-dimensionality of the face, which increases complexity and costs of the system. In this work, we propose an effective and low-cost face spoofing detector system to supplement a PPGI-based vital signal monitoring application. Starting from a set of automatically located facial feature points, the movement information of this set of points was obtained. Based on a time series analysis of the landmark position distances using peak descriptors and cross-correlation coefficients as classifiers in a sliding window, life signs have been exploited to develop a system being able to recognize false detections and biometric spoofs. To verify the performance, experiments on three different benchmark datasets (CASIA face anti-spoofing, MSU and IDIAP Replay-Attack databases) were made. The evaluation of the proposed low-cost approach showed good results (accuracy of ~85-95%) compared to more resource-intensive state-of-the-art methods.

Paper Nr: 26
Title:

Photoplethysmogram Fits Finger Blood Pressure Waveform for non-Invasive and minimally-Intrusive Technologies - Evaluation of Derivative Approaches

Authors:

Gonzalo Tapia, Matias Salinas, Jaime Plaza, Diego Mellado, Rodrigo Salas, Carolina Saavedra, Alejandro Veloz, Alexis Arriola, Juan Idiaquez and Antonio Glaría

Abstract: The purpose of this work is to fit Photoplethysmography (PPG) to finger Arterial Pressure (fiAP) waveform using derivative approaches. Derivative approaches consider using Linear Combination of Derivatives (LCD) and Fractional Derivatives (FDPa). Four informed healthy subjects, aging 35:811:0 years old, agreed to perform Handgrip maneuvers. Signals are recorded continually; a Finapres NOVA device is used for fiAP, while a BIOPAC System is used for PPG and ECG. PPG is smoothed and segmented by heartbeat; recording sections interfered with spiky blocking noise, are eliminated. Finally, PPG is processed using LCD and FDPa and their results are enriched using Lasso technique. Twenty records per subject at rest and twenty at raised BP are analyzed. Results show PPG to fiAP fitting errors 5:38%0:91 at resting fiAP and 5:86%1:21 at raised fiAP, being always lower than 15%, suggesting that derivative approaches could be suitable for medical applications.

Paper Nr: 33
Title:

Frequency Domain Analysis of Acoustic Emission Signals in Medical Drill Wear Monitoring

Authors:

Zrinka Murat, Danko Brezak, Goran Augustin and Dubravko Majetic

Abstract: Medical drills are subject to wear process due to mechanical, thermal and, potentially, sterilisation influences. The influence of drill wear on friction contributes to the drilling temperature rise and occurrence of thermal osteonecrosis. During the cutting process drilling temperature cannot be adequately reduced by applying cooling fluid externally on the bone surface and a part of a tool which is not in the contact with the bone if higher wear rates occurs. Since it is not possible to directly establish or measure drill wear rate without interrupting the machining process, this important parameter should be estimated using available process signals. Therefore, the application of tool wear features extracted from acoustic emission signals in the frequency domain for the purpose of indirect medical drill wear monitoring process has been studied in detail and the results are presented in this paper.

Paper Nr: 34
Title:

Dynamic Bayesian Network Modeling of Hippocampal Subfields Connectivity with 7T fMRI: A Case Study

Authors:

Fernando P. Santos, Stephen F. Smagula, Helmet Karim, Tales S. Santini, Howard J. Aizenstein, Tamer S. Ibrahim and Carlos D. Maciel

Abstract: The development of high resolution structural and functional magnetic resonance imaging, along with the new automatic segmentation procedures for identifying brain regions with high precision and level of detail, has made possible new studies on functional connectivity in the medial temporal lobe and hippocampal subfields, with important applications in the understanding of human memory and psychiatric disorders. Many previous analyses using high resolution data have focused on undirected measures between these subfields. Our work expands this by presenting Dynamic Bayesian Network (DBN) models as an useful tool for mapping directed functional connectivity in the hippocampal subfields. Besides revealing directional connections, DBNs use a model-free approach which also exclude indirect connections between nodes of a graph by means of conditional probability distribution. They also relax the constraint of acyclicity imposed by traditional Bayesian networks (BNs) by considering nodes at different time points through a Markovianity assumption. We apply the GlobalMIT DBN learning algorithm to one subject with fMRI time-series obtained from three regions: the cornu ammonis (CA), dentate gyrus (DG) and entorhinal cortex (ERC), and find an initial network structure, which can be further expanded with the inclusion of new regions and analyzed with a group analysis method.

Paper Nr: 43
Title:

Evaluation of a Dental Caries Clinical Decision Support System

Authors:

Michel Bessani, Daniel Rodrigues de Lima, Emery Cleiton Cabral Correia Lins and Carlos Dias Maciel

Abstract: Decision Support Systems (DSSs) aims to support professionals decision process. A specific area of application is the Clinical one, resulting in Clinical Decision Support Systems (CDSSs), focusing on Clinical Decision problems, like oncology, geriatrics, and dentistry. DSSs integrate expert knowledge through pattern-based approaches. Bayesian Networks are probabilistic graph models that allow representation and inference on complex scenarios. BNs are used in different decision-making fields, e.g., Clinical Decision Support Systems. Traditionally, such models are learned using established databases. However, in situations where such data set is unavailable, the BN can be manually constructed converting expert knowledge in conditional probabilities. In this paper, we evaluate a Dental Caries Clinical Decision Support System which uses a BN to provide suggestions and represent clinical patterns. The evaluation methodology uses forward sampling to generated data from the BN. The generated data are separated into three groups, and each one is analyzed. The results show the certainty of the Bayesian Network for some scenarios. The analysis of the CDSS BN indicates that the system efficiently infers according to the pattern presented in the literature.

Paper Nr: 45
Title:

The Influence of Emotional States on Short-term Memory Retention by using Electroencephalography (EEG) Measurements: A Case Study

Authors:

Ioana A. Badara, Shobhitha Sarab, Abhilash Medisetty, Allen Cook, Joyce Cook and Buket D. Barkana

Abstract: This study explored how emotions can impact short-term memory retention, and thus the process of learning, by analyzing five mental tasks. EEG measurements were used to explore the effects of three emotional states (e.g., neutral, positive, and negative states) on memory retention. The ANT Neuro system with 625Hz sampling frequency was used for EEG recordings. A public-domain library with emotion-annotated images was used to evoke the three emotional states in study participants. EEG recordings were performed while each participant was asked to memorize a list of words and numbers, followed by exposure to images from the library corresponding to each of the three emotional states, and recall of the words and numbers from the list. The ASA software and EEGLab were utilized for the analysis of the data in five EEG bands, which were Alpha, Beta, Delta, Gamma, and Theta. The frequency of recalled event-related words and numbers after emotion arousal were found to be significantly different when compared to those following exposure to neutral emotions. The highest average energy for all tasks was observed in the Delta activity. Alpha, Beta, and Gamma activities were found to be slightly higher during the recall after positive emotion arousal.

Paper Nr: 47
Title:

Identification of Femoral-Acetabular Symptoms using sEMG Signals during Dynamic Contraction

Authors:

Zahra Karimi Tabar, Chris Joslin, Mario Lamontagne and Giulia Mantovani

Abstract: This paper focuses on development of an algorithm that automatically differentiates a Femoro-Acetabular Impingement (FAI) patient from a healthy control person by comparing their surface electromyography (sEMG) signal recorded from Gluteus Maximus (GMax), Tensor Fasciae Latae (TFL), and Rectus Femoris (RF) muscles in the hip area. A discrete wavelet transform (DWT) method was used to analyse sEMG signals by thirty-eight different wavelet functions (WFs) with 5 decomposition levels of dynamic contractions during the three phases (descending, stationary, and ascending) of a squat task. The Bior3.9 WF was selected as it provided higher amount of energy for most of the subjects and then the wavelet power spectrum was computed for healthy control and FAI groups. The results show that the RF muscle is more active in the ascending phase than the descending phase for FAI subjects, whereas it is more active in the descending phase for healthy control. An independent sample t-test was used to check the activities of muscle in both groups. The results demonstrate no significant difference for GMax (p=0.7477) and TFL (p=0.4997) muscles, while there is a significant difference for RF muscle (p=0.0670).

Paper Nr: 54
Title:

Comparison among Voice Activity Detection Methods for Korean Elderly Voice

Authors:

JiYeoun Lee

Abstract: In the elderly voice, a large amount of noise is generated by physiological changes such as respiration, vocalization, and resonance according to age. So it provides a cause for performance degradation when operating a fusion healthcare device such as voice recognition, synthesis, and analysis software with elderly voice. Therefore, it is necessary to analyze and research the voice of elderly people so that they can operate various healthcare devices with their voices. This study investigated the voice activity detection algorithm for the elderly voice using the existing symmetric higher order differential energy function. And it is confirmed that it has better performance in detection of voice interval in the elderly voice compared with the autocorrelation function and average magnitude difference function method. The voice activity detection proposed in this paper can be applied to the voice interface for the elderly, thereby improving the accessibility of the elderly to the smart device. Furthermore, it is expected that the performance improvement and development of various fusion wearable devices for the elderly will be possible.

Paper Nr: 57
Title:

Importance of Sequence Design Methods Considering Hybridization Kinetics for in vivo DNA Computers

Authors:

Toshihiro Kojima and Akira Suyama

Abstract: A DNA computer is a DNA-based synthetic system inspired by biology. One of the goals of DNA computer research is to develop an in vivo DNA computer, which can function within living cells through non-destructively processing intracellular signals under isothermal conditions. DNA computers working in isothermal conditions need a set of nucleotide sequences satisfying a kinetic condition in addition to the thermodynamic conditions considered previously, because the progress of computation under isothermal conditions is often dominated by the rate of nucleic acid hybridization reactions. We thus developed a method to predict the hybridization reaction rate from nucleotide sequences and have demonstrated experimentally the importance of hybridization reaction rates and the usefulness of our method. The present method is general and can be used to develop any hybridization-based DNA/RNA system such as DNA computers, DNA sensors, DNA nanostructures, and nucleic acid drugs, working in isothermal conditions.

Paper Nr: 59
Title:

Semi-supervised Distributed Clustering for Bioinformatics - Comparison Study

Authors:

Huayiing Li and Aleksandar Jeremic

Abstract: Clustering analysis is a widely used technique in bioinformatics and biochemistry for variety of applications such as detection of new cell types, evaluation of drug response, etc. Since different applications and cells may require different clustering algorithms combining multiple clustering results into a consensus clustering using distributed clustering is a popular and efficient method to improve the quality of clustering analysis. Currently existing solutions are commonly based on supervised techniques which do not require any a priori knowledge. However in certain cases, a priori information on particular labelings may be available a priori. In these cases it is expected that performance improvement can be achieved by utilizing this prior information. To this purpose in this paper, we propose two semi-supervised distributed clustering algorithms and evaluate their performance for different base clusterings