Abstract: ABSTRACT: BACKGROUND: Even though temperature is a continuous quantitative variable, its measurement has been considered a snapshot of a process, indicating whether a patient is febrile or afebrile. Recently, other diagnostic techniques have been proposed for the association between different properties of the temperature curve with severity of illness in the Intensive Care Unit (ICU), based on complexity analysis of continuously monitored body temperature. In this study, we tried to assess temperature complexity in patients with systemic inflammation during a suspected ICU-acquired infection, by using wavelets transformation and multiscale entropy of temperature signals, in a cohort of mixed critically ill patients. METHODS: Twenty-two patients were enrolled in the study. In five, systemic inflammatory response syndrome (SIRS, group 1) developed, 10 had sepsis (group 2), and seven had septic shock (group 3). All temperature curves were studied during the first 24 hours of an inflammatory state. A wavelet transformation was applied, decomposing the signal in different frequency components (scales) that have been found to reflect neurogenic and metabolic inputs on temperature oscillations. Wavelet energy and entropy per different scales associated with complexity in specific frequency bands and multiscale entropy of the whole signal were calculated. Moreover, a clustering technique and a linear discriminant analysis (LDA) were applied for permitting pattern recognition in data sets and assessing diagnostic accuracy of different wavelet features among the three classes of patients. RESULTS: Statistically significant differences were found in wavelet entropy between patients with SIRS and groups 2 and 3, and in specific ultradian bands between SIRS and group 3, with decreased entropy in sepsis. Cluster analysis using wavelet features in specific bands revealed concrete clusters closely related with the groups in focus. LDA after wrapper-based feature selection was able to classify with an accuracy of more than 80% SIRS from the two sepsis groups, based on multiparametric patterns of entropy values in the very low frequencies and indicating reduced metabolic inputs on local thermoregulation, probably associated with extensive vasodilatation. CONCLUSIONS: We suggest that complexity analysis of temperature signals can assess inherent thermoregulatory dynamics during systemic inflammation and has increased discriminating value in patients with infectious versus noninfectious conditions, probably associated with severity of illness.
Abstract: Technological advances in textile, biosensor and electrocardiography domain induced the wide spread use of bio-signal acquisition devices leading to the generation of massive bio-signal datasets. Among the most popular bio-signals, electrocardiogram (ECG) possesses the longest tradition in bio-signal monitoring and recording, being a strong and relatively robust signal. As research resources are fostered, research community promotes the need to extract new knowledge from bio-signals towards the adoption of new medical procedures. However, integrated access, query and management of ECGs are impeded by the diversity and heterogeneity of bio-signal storage data formats. In this scope, the proposed work introduces a new methodology for the unified access to bio-signal databases and the accompanying metadata. It allows decoupling information retrieval from actual underlying datasource structures and enables transparent content and context based searching from multiple data resources. Our approach is based on the definition of an interactive global ontology which manipulates the similarities and the differences of the underlying sources to either establish similarity mappings or enrich its terminological structure. We also introduce ROISES (Research Oriented Integration System for ECG Signals), for the definition of complex content based queries against the diverse bio-signal data sources.
Abstract: The present study quantitatively analyzes the EEG characteristics during activations (Act) that occur during NREM sleep, and constitute elements of sleep microstructure (i.e. the Cyclic Alternating Pattern). The fractal dimension (FD) and the sample entropy (SampEn) measures were used to study the different sleep stages and the Act that build up the sleep structure. Polysomnographic recordings from 10 good sleepers were analyzed. The complexity indexes of the Act were compared with the non-activation (NAct) periods during non-REM sleep. In addition, complexity measures among the different Act subtypes (A1, A2 and A3) were analyzed. A3 presented a quite similar complexity independently of the sleep stage, while A1 and A2 showed higher complexity in light sleep than during deep sleep. The current results suggest that Act present a hierarchic complexity between subtypes A3 (higher), A2 (intermediate) and A1 (lower) in all sleep stages.
Abstract: To explore new features of the optic nerve head morphology using the Heidelberg retina tomograph (HRT) and to assess their discriminating power between glaucomatous patients and normal subjects.
Abstract: The aim of the study was to investigate heart rate (HR) and respiratory rate (RR) complexity in patients with weaning failure or success, using both linear and nonlinear techniques.
Abstract: This work investigates the relation between the complexity of electroencephalography (EEG) signal, as measured by fractal dimension (FD), and normal sleep structure in terms of its macrostructure and microstructure. Sleep features are defined, encoding sleep stage and cyclic alternating pattern (CAP) related information, both in short and long term. The relevance of each sleep feature to the EEG FD is investigated, and the most informative ones are depicted. In order to quantitatively assess the relation between sleep characteristics and EEG dynamics, a modeling approach is proposed which employs subsets of the sleep macrostructure and microstructure features as input variables and predicts EEG FD based on these features of sleep micro/macrostructure. Different sleep feature sets are investigated along with linear and nonlinear models. Findings suggest that the EEG FD time series is best predicted by a nonlinear support vector machine (SVM) model, employing both sleep stage/transitions and CAP features at different time scales depending on the EEG activation subtype. This combination of features suggests that short-term and long-term history of macro and micro sleep events interact in a complex manner toward generating the dynamics of sleep.
Abstract: Separation from mechanical ventilation is a difficult task, whereas conventional predictive indices have not been proven accurate enough, so far. A few studies have explored changes of breathing pattern variability for weaning outcome prediction, with conflicting results. In this study, we tried to assess respiratory complexity during weaning trials, using different non-linear methods derived from theory of complex systems, in a cohort of surgical critically ill patients.
Abstract: This work aims to propose new methodologies for the quantitative characterization of insomnia. Sleep microstructure, as expressed by Cyclic Alternatic pattern (CAP) sleep, is studied and differences between normal sleepers and insomniacs are investigated. The dynamic in the structure of CAP activation events is studied by use of wavelet analysis and the content of events, i.e. EEG dynamics, is studied in terms of complexity analysis. Both in structure and content, features exhibiting statistically significant differences are proposed, opening new perspectives for the understanding and the quantitative characterization of sleep and its disorders.
Abstract: HeartCycle is an integrated project aiming to provide a disease management solution for cardiovascular disease patients, by developing technologies, algorithms to interpret data and services to facilitate the remote management of patients at home. In this paper an overview of part of the algorithmic work package, oriented at motivating the patients to be compliant to treatment regimes and to adopt a beneficial lifestyle, will be given. A concept allowing further education of the patient on the effect of medication on their vital signs, as well as the prediction of medication effect and a possible way to check compliance using vital signs measurements will be presented.
Abstract: Exercise constitutes an important intervention aiming to improve health and quality of life for several categories of patients. Personalized exercise prescription is a rather complicated issue, requiring several aspects to be taken into account, e.g. patient's medical history and response to exercise, medication treatment, personal preferences, etc. The present work proposes an ontology-based framework designed to facilitate healthcare professionals in personalized exercise prescription. The framework encapsulates the necessary domain knowledge and the appropriate inference logic, so as to generate exercise plan suggestions based on patient's profile. It also supports readjustments of a prescribed plan according to the patient's response with respect to goal achievement and changes in physical-medical status. An instantiation of the proposed framework for cardiac rehabilitation illustrates the virtue and the applicability of this work.
Abstract: Atrial fibrillation (AF) is a complex phenomenon, related with a multitude of factors, including the electrical properties of the atrial substrate. The purpose of this work is to present a method that highlights electrocardiographic differences between normal subjects and patients with paroxysmal AF episodes (PAF), potentially related with substrate differences. Vectorcardiography recordings are considered and, for each lead (X-Y-Z), on a beat by beat basis, a steady window before QRS, corresponding to the atrial activity, is analysed via continuous wavelet transform. Wavelet-based parameters are calculated and compared between the normal and AF group, with the beat to beat variation of wavelet energy as the most important feature showing a significantly higher variability in the AF group.
Abstract: OBJECTIVES: The pathogenetic mechanisms responsible for the initiation and recurrence of PAF are not fully elucidated and vary among individuals. We evaluated the ability of a novel non-invasive approach based on P wave wavelet analysis to predict symptomatic paroxysmal atrial fibrillation (PAF) recurrences in individuals without structural heart disease. METHODS: We studied 50 patients (24 males, mean age 54.9±9.8years) presented to our emergency department with a symptomatic episode of PAF. The patients were followed-up for 12.1±0.1months and classified into two groups according to the number of PAF episodes: Group A (<5 PAF, n=33), Group B (≥5 PAF, n=17). A third Group of 50 healthy individuals without history of PAF was used as control. Study groups underwent echocardiography and orthogonal ECG-based wavelet analyses of P waves at baseline and follow-up. Maximum and mean P wave energies were calculated in each subject at each orthogonal lead using the Morlet wavelet analysis. RESULTS: Larger P wave energies at X lead and relatively larger left atrium were independently associated with >5 PAF episodes vs. <5 PAF episodes. No difference in P wave duration was detected between Groups A and B (p>0.1), whereas Group A and B patients had longer P waves at Z lead compared to Group C (86.4±13 vs. 71.5±15msec, p<0.001). CONCLUSIONS: P wave wavelet analysis can reliably predict the generation and recurrence of PAF within a year. P wave wavelet analysis could contribute to the early identification of patients at risk for increased number of PAF recurrences.
Abstract: The ongoing efforts toward continuity of care and the recent advances in information and communication technologies have led to a number of successful personal health systems for the management of chronic care. These systems are mostly focused on monitoring efficiently the patient's medical status at home. This paper aims at extending home care services delivery by introducing a novel framework for monitoring the patient's condition and safety with respect to the medication treatment administered. For this purpose, considering a body area network (BAN) with advanced sensors and a mobile base unit as the central communication hub from the one side, and the clinical environment from the other side, an architecture was developed, offering monitoring patterns definition for the detection of possible adverse drug events and the assessment of medication response, supported by mechanisms enabling bidirectional communication between the BAN and the clinical site. Particular emphasis was given on communication and information flow aspects that have been addressed by defining/adopting appropriate formal information structures as well as the service-oriented architecture paradigm. The proposed framework is illustrated via an application scenario concerning hypertension management.
Abstract: Splice sites define the boundaries of exonic regions and dictate protein synthesis and function. The splicing mechanism involves complex interactions among positional and compositional features of different lengths. Computational modeling of the underlying constructive information is especially challenging, in order to decipher splicing-inducing elements and alternative splicing factors. SpliceIT (Splice Identification Technique) introduces a hybrid method for splice site prediction that couples probabilistic modeling with discriminative computational or experimental features inferred from published studies in two subsequent classification steps. The first step is undertaken by a Gaussian support vector machine (SVM) trained on the probabilistic profile that is extracted using two alternative position-dependent feature selection methods. In the second step, the extracted predictions are combined with known species-specific regulatory elements, in order to induce a tree-based modeling. The performance evaluation on human and Arabidopsis thaliana splice site datasets shows that SpliceIT is highly accurate compared to current state-of-the-art predictors in terms of the maximum sensitivity, specificity tradeoff without compromising space complexity and in a time-effective way. The source code and supplementary material are available at: http://www.med.auth.gr/research/spliceit/.
Abstract: Radiofrequency catheter ablation (RFA) is an effective treatment of arrhythmias. However, patients often remain symptomatic after the procedure. We aimed to assess the arrhythmia recurrence after successful RFA in relation to patients' symptoms using transtelephonic loop recorders. Thirty-six consecutive patients (age 50 +/- 14 years, 17 males/19 females) were enrolled after successful RFA for atrioventricular (AV) nodal reentrant tachycardia (n = 21), AV reentrant tachycardia (n = 8), atrial tachycardia (n = 2), atrial fibrillation/flutter (n = 4), and ventricular tachycardia (n = 1). During 23 +/- 6 days of follow-up, 679 events were recorded, 246 of which were true arrhythmic events, mostly (56%) asymptomatic. The vast majority of these true arrhythmic events were due to trivial arrhythmias (extrasystoles or sinus tachycardia), equally distributed among symptomatic and asymptomatic episodes. Arrhythmia relapse was shown in four patients, who had a total of nine episodes, eight of which were symptomatic. No high degree AV block was detected. Overall, symptom recurrence had low sensitivity (44%) and high specificity (95%) for the detection of any arrhythmia, and high sensitivity (89%) but low specificity (58%) for the detection of relapse. In conclusion, transtelephonic monitoring was a useful tool for the assessment of symptoms after RFA and its use may be reserved for the most symptomatic patients to detect a relapse or to reassure them for the benign nature of their symptoms.
Abstract: This work investigates the relation between EEG complexity measures, in particular Fractal Dimension and Sample Entropy, and sleep structure, in terms of both macrostructure, i.e. sleep stages, and microstructure, i.e. phase A activation of CAP sleep. Activation phases are compared with the non-activation periods of non-REM sleep. The study suggests that complexity features can serve as consistent descriptors of sleep dynamics and can potentially assist in the classification of sleep stages.
Abstract: This work aims at examining the influence of lorazepam, an anxiolytic drug with sedative effects, on brain activity and specifically on EEG Fractal Dimension (FD). The main objective is to clarify the reasons for FD increase after drug intake and to establish a relationship between FD and EEG energy bands. 14 healthy subjects that received either 2.5mg of lorazepam (verum case) or placebo (placebo case) participated in the experiment. 20 EEG channels have been used. One-way ANOVA test revealed that lorazepam increases significantly both FD (p = 0) and beta energy band (p = 1.18E-013 for beta1 band and p = 2.29E-011 for beta2 band) and decreases alpha energy band (p = 0.05 for alpha1 and p = 0.0036 for alpha2), whereas there was not any significant difference on placebo subjects before and after drug intake. Moreover, correlation results indicate that there is a strong correlation between FD and beta energy band (mean correlation coefficient = 0.4120 for beta1 and 0.5358 for beta2) and a negative correlation between FD and alpha1 energy band (mean correlation coefficient -0.4930). Additionally, the mean correlation coefficient between FD and a combination of the different energy bands (beta1+beta2-alpha1) is 0.6185 and reaches value 0.684 for channels F7, T6, P4. These results indicate that there is a relationship between EEG energy bands and FD and provide a link between the classic spectral analysis and the complexity analysis.
Abstract: Adverse Drug Events (ADEs) are currently considered as a major public health issue, endangering patients' safety and causing significant healthcare costs. Several research efforts are currently concentrating on the reduction of preventable ADEs by employing Information Technology (IT) solutions, which aim to provide healthcare professionals and patients with relevant knowledge and decision support tools. In this context, we present a knowledge engineering approach towards the construction of a Knowledge-based System (KBS) regarded as the core part of a CDSS (Clinical Decision Support System) for ADE prevention, all developed in the context of the EU-funded research project PSIP (Patient Safety through Intelligent Procedures in Medication). In the current paper, we present the knowledge sources considered in PSIP and the implications they pose to knowledge engineering, the methodological approach followed, as well as the components defining the knowledge engineering framework based on relevant state-of-the-art technologies and representation formalisms.
Abstract: Selecting the most informative features in supervised biological classification problems is a decisive pre-processing step for two main reasons: (1) to deal with the dimensionality reduction problem, and (2) to ascribe biological meaning to the underlying feature interactions. This paper presents a filter-based feature selection method that is suitable for positional modeling of biological sequences. The basic motivation is the problem of using a positional model of fixed length that sub-optimally describes biological sequences in a specific classification problem. The core filtering criterion is the F-score and the source features are the positional probabilities describing variable-length interactions among residues. The proposed method was evaluated on human splice sites classification using a linear SVM classifier. The method yields to superior classification accuracy compared to the individual positional models, while it maintains the space complexity of the individual models, in a time-efficient way and independently of the classifier.
Abstract: Heart Rate Variability (HRV) reflects the balance between sympathetic and parasympathetic activity. Slower HRV rhythms (LF) indicate increased sympathetic and/or lower vagal activity, wakefulness characteristics, while faster HRV rhythms (HF) indicate lower sympathetic and/or increased parasympathetic and vagal activity, sleepy characteristics. In this work we demonstrate that power spectral analysis of drivers' heart rate can report driving errors caused by sleepiness. Furthermore, variation of Fractal Dimension (FD) can aid significant information for the assessment of the driving situation. ECG and EEG data were collected from sleep-deprived subjects exposed to real field driving conditions. A lower ratio of low frequency to high frequency components (LF/HF), and lower LF values were reported on the occurrence of driving errors.
Abstract: Sensor networks constitute the backbone for the construction of personalized monitoring systems. Up to now, several sensor networks have been proposed for diverse pervasive healthcare applications, which are however characterized by a significant lack of open architectures, resulting in closed, non-interoperable and difficult to extend solutions. In this context, we propose an open and reconfigurable wireless sensor network (WSN) for pervasive health monitoring, with particular emphasis in its easy extension with additional sensors and functionality by incorporating embedded intelligence mechanisms.
Abstract: Electrocardiography serves as the gold standard for the noninvasive diagnosis of heart disease. Comparison and review between current and previous ECG recordings and clinical data may considerably leverage the physician's diagnostic accuracy. However, clinical data and ECGs are located in heterogeneous and disparate computer systems. On top of this, the plethora of different and mainly proprietary formats for ECG recording and storing, impedes the efforts towards aggregation of medical information. In this paper we propose an approach towards the seamless integration of and access to disparate and heterogeneous sources of medical information, through a semantically enhanced platform called SCP-ECG Integrated Access (SEIA). SEIA provides the means to contextualize SCP-ECG biosignals, by semantically structuring the clinical information they contain. The semantic annotation procedure is enhanced with further processing of the biosignal, including waveform recognition and feature extraction, for automatic or semiautomatic insertion of annotations where missing. Diagnosis and evaluation of a patient's clinical condition is facilitated through integrated temporal comparison of the medical information derived from successive ECGs, in terms of their features and annotations, along with clinical data and measurements originating from the Electronic Medical Record (EMR).
Abstract: A flexible, scaleable and cost-effective medical telemetry system is described for monitoring sleep-related disorders in the home environment. The system was designed and built for real-time data acquisition and processing, allowing for additional use in intensive care unit scenarios where rapid medical response is required in case of emergency. It comprises a wearable body area network of Zigbee-compatible wireless sensors worn by the subject, a central database repository residing in the medical centre and thin client workstations located at the subject's home and in the clinician's office. The system supports heterogeneous setup configurations, involving a variety of data acquisition sensors to suit several medical applications. All telemetry data is securely transferred and stored in the central database under the clinicians' ownership and control.
Abstract: The objective of this study is the development and evaluation of efficient neurophysiological signal statistics, which may assess the driver's alertness level and serve as potential indicators of sleepiness in the design of an on-board countermeasure system.
Abstract: The problem addressed in this work is sleepiness during driving, which often leads to accidents in the streets. Experiments with sleepy drivers took place and the EEG data were analysed in terms of non-linear methods. Sample entropy and phase synchronization variations were investigated within the signal sections corresponding to "driving events", i.e. driving mistakes or loss of control, as well as to periods of drowsiness and sleepiness, as compared to the periods of normal driving. Decreased sample entropy, indicating loss of complexity, and an increased phase synchronisation have been found in the preliminary study presented. The results are encouraging towards developing an alerting system for predicting and preventing driving accidents.
Abstract: For the purposes of insomnia treatment, pharmacotherapy is widely used, despite the possibility for the use of behavioural treatment of insomnia. Thus, the assessment and treatment of patients with insomnia needs further investigation. This work addresses insomnia treatment evaluation and medication side-effect assessment based on continuous physiological signals such as EEG and ECG monitoring and analysis. EEG and ECG measurements regarding drug medication (verum/placebo cases) have been used in a series of experiments, where spectral and non-linear features have been calculated, for assessing a possible distinct behaviour between the verum/placebo condition and furthermore the relation of features to a physiological conditions. Results show that a combination of EEG and ECG based characteristics, both spectral and non-linear, can be used to reveal the differences introduced with insomnia medication treatment, either being improvement in the hyperarousal state, or undesired side effects.
Abstract: To investigate longitudinally over time heart rate dynamics and relation with mortality and organ dysfunction alterations in patients admitted to a multidisciplinary intensive care unit.
Abstract: Driver sleepiness due to sleep deprivation is a causative factor in 1% to 3% of all motor vehicle crashes. In recent studies, the importance of developing driver fatigue countermeasure devices has been stressed, in order to help prevent driving accidents and errors. Although numerous physiological indicators are available to describe an individual's level of alertness, the EEG signal has been shown to be one of the most predictive and reliable, since it is a direct measure of brain activity. In the present study, multichannel EEG data that were collected from 20 sleep-deprived subjects during real environmental conditions of driving are presented for the first time. EEG data's annotation made by two independent Medical Doctors revealed an increase of slowing activity and an acute increase of the alpha waves 5-10 seconds before driving events. From the EEG data that were collected, the Relative Band Ratio (RBR) of the EEG frequency bands, the Shannon Entropy, and the Kullback-Leibler (KL) Entropy were estimated for each one second segment. The mean values of these measurements were estimated for 5 minutes periods. Analysis revealed a significant increase of alpha waves relevant band ratios (RBR), a decrease of gamma waves RBR, and a significant decrease of KL entropy when the first and the last 5-min periods were compared. A rapid decrease of both Shannon and K-L entropies was observed just before the driving events. Conclusively, EEG can assess effectively the brain activity alterations that occur a few seconds before sleeping/drowsiness events in driving, and its quantitative measurements can be used as potential sleepiness indicators for future development of driver fatigue countermeasure devices.
Abstract: Biomedical applications are becoming increasingly reliant on resource integration and information exchange within global solution frameworks that offer seamless connectivity and data sharing in distributed environments. Resource autonomy and data heterogeneity are the most important impediments towards this potential. Aiming to overcome these limitations, we propose an implementation of the service-oriented model towards the construction of an open, semantically enriched biomedical service space that enables advanced service registration, selection and access capabilities, as well as service interoperability. The proposed system is realised by defining service annotation ontologies and applying software agent technology as the means for service registration, matchmaking and interfacing in a Grid environment. The applicability of the envisioned biomedical service space is illustrated on a set of bioinformatics resources, addressing computational identification of protein-coding genes.
Abstract: Over the last four years, a community of researchers working on Grid and High Performance Computing technologies started discussing the barriers and opportunities that grid technologies must face and exploit for the development of health-related applications. This interest lead to the first Healthgrid conference, held in Lyon, France, on January 16th-17th, 2003, with the focus of creating increased awareness about the possibilities and advantages linked to the deployment of grid technologies in health, ultimately targeting the creation of a European/international grid infrastructure for health. The topics of this conference converged with the position of the eHealth division of the European Commission, whose mandate from the Lisbon Meeting was "To develop an intelligent environment that enables ubiquitous management of citizens' health status, and to assist health professionals in coping with some major challenges, risk management and the integration into clinical practice of advances in health knowledge." In this context "Health" involves not only clinical procedures but covers the whole range of information from molecular level (genetic and proteomic information) over cells and tissues, to the individual and finally the population level (social healthcare). Grid technology offers the opportunity to create a common working backbone for all different members of this large "health family" and will hopefully lead to an increased awareness and interoperability among disciplines. The first HealthGrid conference led to the creation of the Healthgrid association, a non-profit research association legally incorporated in France but formed from the broad community of European researchers and institutions sharing expertise in health grids. After the second Healthgrid conference, held in Clermont-Ferrand on January 29th-30th, 2004, the need for a "white paper" on the current status and prospective of health grids was raised. Over fifty experts from different areas of grid technologies, eHealth applications and the medical world were invited to contribute to the preparation of this document.
Abstract: Current advances in biosensor technology allow multiple miniaturized or textile sensors to record continuously biosignals, such as blood pressure or heart rate, and transmit the information of interest to clinical sites. New applications are emerging, based on such systems, towards pervasive healthcare. This paper describes an architecture enabling biosensors, forming a Body Area Network (BAN), to be integrated in a Grid infrastructure. The Grid services proposed, such as access to recorded data, are offered via the BAN console, an enhanced wearable computer, where the recordings of multiple biosensors are integrated. Medical Grid-enabled Nodes can have access to biosensor measurements upon demand, or can agree to get notifications and alerts. Thus, in such a distributed environment, data and computational resources are independent, yet cooperating unobtrusively, contributing to the notion of pervasive healthcare.
Abstract: In this paper, a multiagent system (MAS) is presented, aiming to enhance monitoring, surveillance, and educational services of a generic medical contact center (MCC) for chronic disease management. In such a home-care scenario, a persistent need arises for efficiently monitoring the patient contacts and the MCC's functionality, in order to effectively manage and interpret the large volume of medical data collected during the patient sessions with the system, and to assess the use of MCC resources. Software agents were adopted to provide the means to accomplish such real-time information-processing tasks, due to their autonomous, reactive and/or proactive nature, and their effectiveness in dynamic environments by incorporating coordination strategies. Specifically, the objective of the MAS is to monitor the MCC environment, detect important cases, and inform the healthcare and administrative personnel via alert messages, notifications, recommendations, and reports, prompting them for actions. The main aim of this paper is to present the overall design and implementation of a proposed MAS, emphasizing its functional model and architecture, as well as on the agent interactions and the knowledge-sharing mechanism incorporated, in the context of a generic MCC.
Abstract: In the context of the Citizen Health System (CHS) project, a modular Medical Contact Center (MCC) was developed, which can be used in the monitoring, treatment, and management of chronically ill patients at home, such as diabetic or congestive heart failure patients. The virtue of the CHS contact center is that, using any type of communication and telematics technology, it is able to provide timely and preventive prompting to the patients, thus, achieving better disease management. In this paper, we present the structure of the CHS system, describing the modules that enable its flexible and extensible architecture. It is shown, through specific examples, how quality of healthcare delivery can be increased by using such a system.
Abstract: Current approaches for mining association rules usually assume that the mining is performed in a static database, where the problem of missing attribute values does not practically exist. However, these assumptions are not preserved in some medical databases, like in a home care system. In this paper, a novel uncertainty rule algorithm is illustrated, namely URG-2 (Uncertainty Rule Generator), which addresses the problem of mining dynamic databases containing missing values. This algorithm requires only one pass from the initial dataset in order to generate the item set, while new metrics corresponding to the notion of Support and Confidence are used. URG-2 was evaluated over two medical databases, introducing randomly multiple missing values for each record's attribute (rate: 5-20% by 5% increments) in the initial dataset. Compared with the classical approach (records with missing values are ignored), the proposed algorithm was more robust in mining rules from datasets containing missing values. In all cases, the difference in preserving the initial rules ranged between 30% and 60% in favour of URG-2. Moreover, due to its incremental nature, URG-2 saved over 90% of the time required for thorough re-mining. Thus, the proposed algorithm can offer a preferable solution for mining in dynamic relational databases.
Abstract: The aim of the present paper, is the estimation of the distance between an electrode used as a recording site of the extracellular potential field and a surviving myocardial bundle. The importance of the reliable solution of this problem lies among others in controlling ablation. For our purposes one-dimensional propagation is considered and current sources are activated along a cable simulating the propagating waves with constant velocity. Different models of current sources are explored. By use of these models, the corresponding functions expressing extracellular potentials are calculated, using the volume conductor equation. This way, extracellular potentials are modeled as parametric functions of longitudinal distance, while perpendicular distance, current source strength, and other factors related to the propagated wave are parameters of the functions. Simulated annealing is applied for model parameter estimation and appropriate Time Domain and Wavelet Domain cost functions are investigated. Different combinations of model and cost function are evaluated regarding the accuracy of distance estimation. A continuous source model function with a wavelet cost function was found to be the most accurate combination. The accuracy of distance estimation is related to the selected source model and to the actual distance of recording in a nonmonotonic way.
Abstract: The purpose of this study was the evaluation of Morlet wavelet analysis of the P wave as a means of predicting the development of atrial fibrillation (AF) in patients who undergo coronary artery bypass grafting (CABG). The P wave was analyzed using the Morlet wavelet in 50 patients who underwent successful CABG. Group A consisted of 17 patients, 12 men and 5 women, of mean age 66.9 +/- 5.9 years, who developed AF postoperatively. Group B consisted of 33 patients, 29 men and 4 women, mean age 62.4 +/- 7.8 years, who remained arrhythmid-free. Using custom-designed software, P wave duration and wavelet parameters expressing the mean and maximum energy of the P wave were calculated from 3-channel digital recordings derived from orthogonal ECG leads (X, Y, and Z), and the vector magnitude (VM) was determined in each of 3 frequency bands (200-160 Hz, 150-100 Hz and 90-50 Hz). Univariate logistic-regression analysis identified a history of hypertension, the mean and maximum energies in all frequency bands along the Z axis, the mean and maximum energies (expressed by the VM) in the 200-160 Hz frequency band, and the mean energy in the 150-100 Hz frequency band along the Y axis as predictors for post-CABG AF. Multivariate analysis identified hypertension, ejection fraction, and the maximum energies in the 90-50 Hz frequency band along the Z and composite-vector axes as independent predictors. This multivariate model had a sensitivity of 91% and a specificity of 65%. We conclude that the Morlet wavelet analysis of the P wave is a very sensitive method of identifying patients who are likely to develop AF after CABG. The occurrence of post-CABG AF can be explained by a different activation pattern along the Z axis.
Abstract: Non-stationary analysis of electrocardiograms (ECGs) using Wigner-Ville distribution is presented. Analysis was performed on subjects with acute myocardial infarction who had undergone thrombolysis, in Holter recordings of lead V1. The distinction between successfully and non-successfully thrombolysed patients was evaluated, based on time-frequency features of the Wigner-Ville transformed ECGs at the sixth hour after lysis. Characteristic parameters were extracted from time-frequency areas, and linear discriminant analysis was performed on these parameters, leading to a prediction index to distinguish the two classes. Thirteen features were found statistically significant by t-test and were used for the classification with linear modelling. Out of these features, four corresponded to frequencies lower than 25 Hz and higher than 50 Hz for, roughly, the QRS complex, five features corresponded to all the frequency bands of, roughly, the ST area, and the last four features corresponded to the T-wave. The feature-vector used in linear modelling was iteratively generated, and the iterative prediction found all 18 features significant. The iterative method resulted in better classification than that of the standard statistical procedure (3.8% error against 18.1% with the classic method). The evolution of the prediction index with time for the first 12 h was different for the successfully and non-successfully thrombolysed groups. Specifically, in the successful thrombolysis group, oscillations and variation with time were more obvious, indicating a possible difference in the dynamics of the cardiac system.
Abstract: In the context of an IST European project with acronym PANACEIA-ITV, a home care service provisioning system is described, based on interactive TV technology. The purpose of PANACEIA-ITV is to facilitate essential lifestyle changes and to promote compliance with scientifically sound self-care recommendations, through the application of interactive digital television for family health maintenance. The means to achieve these goals are based on technological, health services and business models. PANACEIA-ITV is looking for communication of monitoring micro-devices with I-TV set-top-boxes using infrared technology, and embodiment of analogous H/W and S/W in the I-TV set-top-boxes. Intelligent agents are used to regulate data flow, user queries as well as service provisions from and to the household through the satellite digital platform, the portal and the back-end decision support mechanisms, using predominantly the Active Service Provision (ASP) model. Moreover, interactive digital TV services are developed for the delivery of health care in the home care environment.
Abstract: In the present paper, a multi-agent system is proposed, which can be integrated in the home care telemedicine system that was developed in the context of the Citizen Health System (CHS) European project, functioning as a contact center for diabetic and congestive heart failure patients. The objective of the multi-agent system is to provide a set of alert/notification mechanisms for the clinicians, helping them to classify the clinical condition of each patient. Therefore, despite the huge amount of data managed by the system, due to the daily use of the contact center's services, these alert mechanisms provide the clinician with an overview of the cases that need further examination and save him/her time from the trivial cases. The multi-agent system consists of different types of agents, each one assigned with specific tasks, which communicate with each other, in order to share knowledge.
Abstract: Health delivery practices are shifting towards home care. The reasons are the better possibilities for managing chronic care, controlling health delivery costs, increasing quality of life and quality of health services and the distinct possibility of predicting and thus avoiding serious complications. For the above goals to become routine, new telemedicine and information technology (IT) solutions need to be implemented and integrated in the health delivery scene, and these solutions need to be assessed through evidence-based medicine in order to provide solid proof for their usefulness. Thus, the concept of contact or call centers has emerged as a new and viable reality in the field of IT for health and telemedicine. In this paper we describe a generic contact center that was designed in the context of an EU funded IST for health project with acronym Citizen Health System (CHS). Since the generic contact center is composed by a number of modules, we shall concentrate in the modules dealing with the communication between the patient and the contact center using mobile telecommunications solutions, which can act as link between the internet and the classical computer telephony communication means. We further elaborate on the development tools of such solutions, the interface problems we face, and on the means to convey information from and to the patient in an efficient and medically acceptable way. This application proves the usefulness of wireless technology in providing health care services all around the clock and everywhere the citizen is located, it proves the necessity for restructuring the medical knowledge for education delivery to the patient, and it shows the virtue of interactivity by means of using the limited, yet useful browsing capabilities of the wireless application protocol (WAP) technology.
Abstract: a) The use of information technology (IT) based solutions for quality health delivery in regional health information networks and the study of the enabling factors for their use in a regional health care network from key classes of users such as the medical personnel and the citizens. b) Identification of potential technologies for usage from all citizens and health providers in a regional environment, in all aspects of everyday life. c) Presentation of a generic user model for reference when developing and assessing IT based health delivery solutions.
Abstract: The Citizen Health System (CHS) is a European Commission (EC) funded project in the field of IST for Health. Its main goal is to develop a generic contact center which in its pilot stage can be used in the monitoring, treatment and management of chronically ill patients at home in Greece, Spain and Germany. Such contact centers, which can use any type of communication technology, and can provide timely and preventive prompting to the patients are envisaged in the future to evolve into well-being contact centers providing services to all citizens. In this paper, we present the structure of such a generic contact center and in particular the telecommunication infrastructure, the communication protocols and procedures, and finally the educational modules that are integrated into this contact center. We discuss the procedures followed for two target groups of patients where two randomized control clinical trials are under way, namely diabetic patients with obesity problems, and congestive heart failure patients. We present examples of the communication means between the contact center medical personnel and these patients, and elaborate on the educational issues involved.
Abstract: In infarcted myocardium, extracellular recordings exhibit multiple deflections due to irregular pathway of the electric impulse. In this work the problem of distinguishing local from distant deflections is tackled. In order to evaluate the proposed methods in a controlled setting, simulated data are used, following both Beeler-Reuter and Luo-Rudy kinetics. The input is an array of electrograms positioned on grid-points of a rectangular grid and the output is an array of estimates of the membrane current. First, deconvolution techniques are used in the form of spatial filtering for membrane current estimation from the extracellular recordings. Second, the extracellular recordings undergo wavelet based transformation, followed by a spatial filter which enhances local activity deflections and suppresses distant activity deflections. It is shown that wavelet filtering of the extracellular recordings acts as an evaluator of the efficiency of the deconvolution techniques for the membrane current estimation. Subsequently, activation times based on the results from the two methods are used for the reconstruction of the propagation pattern in a zig-zag case in two-dimensional grids. It is shown that the wavelet-based method is more robust, and can work well even in cases where the grid interval in the y direction is four times larger than the single cell size.
Abstract: Health care delivery is changing drastically. In its current state it tends to use the home care model in order to increase quality of life, to rationalize costs and to achieve wellness. Pivotal to these purposes are contact centers, which act as mediators between the medical staff and the citizens seeking advice and/or therapy. Main platforms used for the development of such applications are the INTERNET and PCs, and the telecommunication networks, including mobile solutions. In this paper, a generic contact center model shall be presented, which is under development in the context of an IST European project in health telematics entitled â Distance Information Technologies for Home care. The Citizen Health System (CHS)'. After the description of this generic contact center, an application for health care delivery to diabetic patients shall be described. In this application we shall see the possible use of the Wireless Application Protocol (WAP) scheme. This application proves the usefulness of wireless technology in providing health care services all around the clock and everywhere the citizen is located, it shows the necessity for restructuring the medical knowledge for education delivery to the patient, and it shows the virtue of interactivity by means of using the limited, yet useful browsing capabilities of the WAP technology.
Abstract: The purpose of this work is the enhancement of local activation in multiphasic extracellular recordings coming from infarcted myocardial tissue and, consequently, the distinction of local from distal activation. Deconvolution procedures are applied, in the form of spatial filtering, in order to estimate transmembrane currents from the extracellular recordings. Simulated data are used in order to test the methods. The current source estimates are compared to the actual transmembrane currents and to Laplacian estimates.
Abstract: The huge amount of information involved in clinical cardiological examination raises the need for efficient patient data management and for the fusion of modern information processing techniques in the everyday clinical workstation. An integrated medical information system has been developed in order to organize the local cardiological legacy system in the AHEPA hospital. An ODBC based database (like Ms SqlServer or MSAccess) holds patient and ECG data. The system incorporates data management (storing and retrieval) and data processing modules. The processing module is added as an independent DLL. The visualization component results in a better view of the information. The components are integrated in a friendly window interface that lets the doctor browse patient information, apply modern signal processing techniques, make special measurements and store them in the database for research reasons.
Abstract: Aim of this study is to explore and quantify possible beat-to-beat variations in P wave characteristics. Two minute X-lead ECG recordings were analyzed. Data were obtained from Physionet database. Each P wave was fitted by a Gaussian function whose parameter variations were investigated over two minutes window. The normalized root mean squared error (NRMSE) between each P wave and its fitting was computed. To evaluate the variability of the estimated parameters, the coefficient of variation (CV) was computed. The proposed model well fitted the P waves, being the mean NRMSE 0.047 ± 0.030 mV (range 0.012 - 0.140 mV). A variability was found in all patients for both the parameters A and σ, reflecting P waves amplitude and duration, respectively, being CVA 0.087 ± 0.050 mV (range 0.032 - 0.253 mV) and CV 0.093 ± 0.053 (range 0.028 - 0.27). These preliminary results could be used as a starting point for future beat to-beat P wave analysis in patients with atrial conduction pathologies.
