Service-oriented architecture for device lifecycle support in industrial automation

Dissertação para obtenção do Grau de Doutor em Engenharia Electrotécnica e de Computadores Especialidade: Robótica e Manufactura Integrada

Service-oriented infrastructure to support the control, monitoring and management of a shop floor system

Dissertação para obtenção do Grau de Mestre em Engenharia Electrotécnica e de Computadores

Intelligent decision support to predict patient barotrauma risk in intensive care units

The occurrence of Barotrauma is identified as a major concern for health professionals, since it can be fatal for patients. In order to support the decision process and to predict the risk of occurring barotrauma Data Mining models were induced. Based on this principle, the present study addresses the Data Mining process aiming to provide hourly probability of a patient has Barotrauma. The process of discovering implicit knowledge in data collected from Intensive Care Units patientswas achieved through the standard process Cross Industry Standard Process for Data Mining. With the goal of making predictions according to the classification approach they several DM techniques were selected: Decision Trees, Naive Bayes and Support Vector Machine. The study was focused on identifying the validity and viability to predict a composite variable. To predict the Barotrauma two classes were created: “risk” and “no risk”. Such target come from combining two variables: Plateau Pressure and PCO2. The best models presented a sensitivity between 96.19% and 100%. In terms of accuracy the values varied between 87.5% and 100%. This study and the achieved results demonstrated the feasibility of predicting the risk of a patient having Barotrauma by presenting the probability associated.

Different module placements in a modular façade system for natural ventilation

Nowadays natural ventilation has gained prominence because its correct use can reduce energy consumption for cooling systems and improve thermal comfort among users. In this paper, we report on the modelling initiative, based on the wind tunnel tests that were carried out for the determination of the influence of natural ventilation in buildings. Indeed, the renewal of air in a closed environment without using an air conditioning system with mechanical elements can lead to energy savings and, in addition, provide air quality.The wind tunnel tests were carried out by varying the positioning of six ventilation modules in the façade system configuration. The modules were positioned below the window-sill (ventilated window-sill) as well as separately above and below the façade. The wind speed measurements were taken inside and outside the model for the different façades configurations to evaluate the best performance in relation to natural ventilation. The results supported the positioning of the six ventilation modules below the window-sill, forming a â ventilated window-sillâ as the most effective natural ventilation solution.

Oxygen therapy, continuous positive airway pressure, or noninvasive bilevel positive pressure ventilation in the treatment of acute cardiogenic pulmonary edema

OBJECTIVE: To compare the effects of 3 types of noninvasive respiratory support systems in the treatment of acute pulmonary edema: oxygen therapy (O2), continuous positive airway pressure, and bilevel positive pressure ventilation. METHODS: We studied prospectively 26 patients with acute pulmonary edema, who were randomized into 1 of 3 types of respiratory support groups. Age was 69±7 years. Ten patients were treated with oxygen, 9 with continuous positive airway pressure, and 7 with noninvasive bilevel positive pressure ventilation. All patients received medicamentous therapy according to the Advanced Cardiac Life Support protocol. Our primary aim was to assess the need for orotracheal intubation. We also assessed the following: heart and respiration rates, blood pressure, PaO2, PaCO2, and pH at begining, and at 10 and 60 minutes after starting the protocol. RESULTS: At 10 minutes, the patients in the bilevel positive pressure ventilation group had the highest PaO2 and the lowest respiration rates; the patients in the O2 group had the highest PaCO2 and the lowest pH (p<0.05). Four patients in the O2 group, 3 patients in the continuous positive pressure group, and none in the bilevel positive pressure ventilation group were intubated (p<0.05). CONCLUSION: Noninvasive bilevel positive pressure ventilation was effective in the treatment of acute cardiogenic pulmonary edema, accelerated the recovery of vital signs and blood gas data, and avoided intubation.

Pervasive adaptive data acquisition gateway for critical healthcare

The data acquisition process in real-time is fundamental to provide appropriate services and improve health professionals decision. In this paper a pervasive adaptive data acquisition architecture of medical devices (e.g. vital signs, ventilators and sensors) is presented. The architecture was deployed in a real context in an Intensive Care Unit. It is providing clinical data in real-time to the INTCare system. The gateway is composed by several agents able to collect a set of patients’ variables (vital signs, ventilation) across the network. The paper shows as example the ventilation acquisition process. The clients are installed in a machine near the patient bed. Then they are connected to the ventilators and the data monitored is sent to a multithreading server which using Health Level Seven protocols records the data in the database. The agents associated to gateway are able to collect, analyse, interpret and store the data in the repository. This gateway is composed by a fault tolerant system that ensures a data store in the database even if the agents are disconnected. The gateway is pervasive, universal, and interoperable and it is able to adapt to any service using streaming data.

Optimization techniques to detect early ventilation extubation in intensive care units

The decision support models in intensive care units are developed to support medical staff in their decision making process. However, the optimization of these models is particularly difficult to apply due to dynamic, complex and multidisciplinary nature. Thus, there is a constant research and development of new algorithms capable of extracting knowledge from large volumes of data, in order to obtain better predictive results than the current algorithms. To test the optimization techniques a case study with real data provided by INTCare project was explored. This data is concerning to extubation cases. In this dataset, several models like Evolutionary Fuzzy Rule Learning, Lazy Learning, Decision Trees and many others were analysed in order to detect early extubation. The hydrids Decision Trees Genetic Algorithm, Supervised Classifier System and KNNAdaptive obtained the most accurate rate 93.2%, 93.1%, 92.97% respectively, thus showing their feasibility to work in a real environment.

Multi-scale support vector algorithms for hot spot detection and modelling

The algorithmic approach to data modelling has developed rapidly these last years, in particular methods based on data mining and machine learning have been used in a growing number of applications. These methods follow a data-driven methodology, aiming at providing the best possible generalization and predictive abilities instead of concentrating on the properties of the data model. One of the most successful groups of such methods is known as Support Vector algorithms. Following the fruitful developments in applying Support Vector algorithms to spatial data, this paper introduces a new extension of the traditional support vector regression (SVR) algorithm. This extension allows for the simultaneous modelling of environmental data at several spatial scales. The joint influence of environmental processes presenting different patterns at different scales is here learned automatically from data, providing the optimum mixture of short and large-scale models. The method is adaptive to the spatial scale of the data. With this advantage, it can provide efficient means to model local anomalies that may typically arise in situations at an early phase of an environmental emergency. However, the proposed approach still requires some prior knowledge on the possible existence of such short-scale patterns. This is a possible limitation of the method for its implementation in early warning systems. The purpose of this paper is to present the multi-scale SVR model and to illustrate its use with an application to the mapping of Cs137 activity given the measurements taken in the region of Briansk following the Chernobyl accident.

Patient-ventilator asynchrony during non-invasive ventilation for acute respiratory failure: a multicenter study.

OBJECTIVE : To determine the prevalence of patient-ventilator asynchrony in patients receiving non-invasive ventilation (NIV) for acute respiratory failure. DESIGN : Prospective multicenter observation study. SETTING : Intensive care units in three university hospitals. METHODS: Patients consecutively admitted to ICU were included. NIV, performed with an ICU ventilator, was set by the clinician. Airway pressure, flow, and surface diaphragmatic electromyography were recorded continuously for 30 min. Asynchrony events and the asynchrony index (AI) were determined from visual inspection of the recordings and clinical observation. RESULTS: A total of 60 patients were included, 55% of whom were hypercapnic. Auto-triggering was present in 8 (13%) patients, double triggering in 9 (15%), ineffective breaths in 8 (13%), premature cycling 7 (12%) and late cycling in 14 (23%). An AI > 10%, indicating severe asynchrony, was present in 26 patients (43%), whose median (25-75 IQR) AI was 26 (15-54%). A significant correlation was found between the magnitude of leaks and the number of ineffective breaths and severity of delayed cycling. Multivariate analysis indicated that the level of pressure support and the magnitude of leaks were weakly, albeit significantly, associated with an AI > 10%. Patient comfort scale was higher in pts with an AI < 10%. CONCLUSION: Patient-ventilator asynchrony is common in patients receiving NIV for acute respiratory failure. Our results suggest that leaks play a major role in generating patient-ventilator asynchrony and discomfort, and point the way to further research to determine if ventilator functions designed to cope with leaks can reduce asynchrony in the clinical setting.

Adaptive filtering methods for identifying cross-frequency couplings in human EEG.

Oscillations have been increasingly recognized as a core property of neural responses that contribute to spontaneous, induced, and evoked activities within and between individual neurons and neural ensembles. They are considered as a prominent mechanism for information processing within and communication between brain areas. More recently, it has been proposed that interactions between periodic components at different frequencies, known as cross-frequency couplings, may support the integration of neuronal oscillations at different temporal and spatial scales. The present study details methods based on an adaptive frequency tracking approach that improve the quantification and statistical analysis of oscillatory components and cross-frequency couplings. This approach allows for time-varying instantaneous frequency, which is particularly important when measuring phase interactions between components. We compared this adaptive approach to traditional band-pass filters in their measurement of phase-amplitude and phase-phase cross-frequency couplings. Evaluations were performed with synthetic signals and EEG data recorded from healthy humans performing an illusory contour discrimination task. First, the synthetic signals in conjunction with Monte Carlo simulations highlighted two desirable features of the proposed algorithm vs. classical filter-bank approaches: resilience to broad-band noise and oscillatory interference. Second, the analyses with real EEG signals revealed statistically more robust effects (i.e. improved sensitivity) when using an adaptive frequency tracking framework, particularly when identifying phase-amplitude couplings. This was further confirmed after generating surrogate signals from the real EEG data. Adaptive frequency tracking appears to improve the measurements of cross-frequency couplings through precise extraction of neuronal oscillations.

Neurally adjusted ventilatory assist (NAVA) improves patient-ventilator interaction during non-invasive ventilation delivered by face mask.

PURPOSE: To determine if, compared to pressure support (PS), neurally adjusted ventilatory assist (NAVA) reduces patient-ventilator asynchrony in intensive care patients undergoing noninvasive ventilation with an oronasal face mask. METHODS: In this prospective interventional study we compared patient-ventilator synchrony between PS (with ventilator settings determined by the clinician) and NAVA (with the level set so as to obtain the same maximal airway pressure as in PS). Two 20-min recordings of airway pressure, flow and electrical activity of the diaphragm during PS and NAVA were acquired in a randomized order. Trigger delay (T(d)), the patient's neural inspiratory time (T(in)), ventilator pressurization duration (T(iv)), inspiratory time in excess (T(iex)), number of asynchrony events per minute and asynchrony index (AI) were determined. RESULTS: The study included 13 patients, six with COPD, and two with mixed pulmonary disease. T(d) was reduced with NAVA: median 35 ms (IQR 31-53 ms) versus 181 ms (122-208 ms); p = 0.0002. NAVA reduced both premature and delayed cyclings in the majority of patients, but not the median T(iex) value. The total number of asynchrony events tended to be reduced with NAVA: 1.0 events/min (0.5-3.1 events/min) versus 4.4 events/min (0.9-12.1 events/min); p = 0.08. AI was lower with NAVA: 4.9 % (2.5-10.5 %) versus 15.8 % (5.5-49.6 %); p = 0.03. During NAVA, there were no ineffective efforts, or late or premature cyclings. PaO(2) and PaCO(2) were not different between ventilatory modes. CONCLUSION: Compared to PS, NAVA improved patient ventilator synchrony during noninvasive ventilation by reducing T(d) and AI. Moreover, with NAVA, ineffective efforts, and late and premature cyclings were absent.

Performance of an ICU ventilator and two turbin-based ventilators dedicated to non invasive ventilation (NIV) in simulated high inspiratory effort and rate: a NIV-bench-study.

INTRODUCTION. The role of turbine-based NIV ventilators (TBV) versus ICU ventilators with NIV mode activated (ICUV) to deliver NIV in case of severe respiratory failure remains debated. OBJECTIVES. To compare the response time and pressurization capacity of TBV and ICUV during simulated NIV with normal and increased respiratory demand, in condition of normal and obstructive respiratory mechanics. METHODS. In a two-chamber lung model, a ventilator simulated normal (P0.1 = 2 mbar, respiratory rate RR = 15/min) or increased (P0.1 = 6 mbar, RR = 25/min) respiratory demand. NIV was simulated by connecting the lung model (compliance 100 ml/mbar; resistance 5 or 20 l/mbar) to a dummy head equipped with a naso-buccal mask. Connections allowed intentional leaks (29 ± 5 % of insufflated volume). Ventilators to test: Servo-i (Maquet), V60 and Vision (Philips Respironics) were connected via a standard circuit to the mask. Applied pressure support levels (PSL) were 7 mbar for normal and 14 mbar for increased demand. Airway pressure and flow were measured in the ventilator circuit and in the simulated airway. Ventilator performance was assessed by determining trigger delay (Td, ms), pressure time product at 300 ms (PTP300, mbar s) and inspiratory tidal volume (VT, ml) and compared by three-way ANOVA for the effect of inspiratory effort, resistance and the ventilator. Differences between ventilators for each condition were tested by oneway ANOVA and contrast (JMP 8.0.1, p\0.05). RESULTS. Inspiratory demand and resistance had a significant effect throughout all comparisons. Ventilator data figure in Table 1 (normal demand) and 2 (increased demand): (a) different from Servo-i, (b) different from V60.CONCLUSION. In this NIV bench study, with leaks, trigger delay was shorter for TBV with normal respiratory demand. By contrast, it was shorter for ICUV when respiratory demand was high. ICUV afforded better pressurization (PTP 300) with increased demand and PSL, particularly with increased resistance. TBV provided a higher inspiratory VT (i.e., downstream from the leaks) with normal demand, and a significantly (although minimally) lower VT with increased demand and PSL.

Neonatal and adult ICU ventilators to provide ventilation in neonates, infants, and children: a bench model study.

BACKGROUND: Using a bench test model, we investigated the hypothesis that neonatal and/or adult ventilators equipped with neonatal/pediatric modes currently do not reliably administer pressure support (PS) in neonatal or pediatric patient groups in either the absence or presence of air leaks. METHODS: PS was evaluated in 4 neonatal and 6 adult ventilators using a bench model to evaluate triggering, pressurization, and cycling in both the absence and presence of leaks. Delivered tidal volumes were also assessed. Three patients were simulated: a preterm infant (resistance 100 cm H2O/L/s, compliance 2 mL/cm H2O, inspiratory time of the patient [TI] 400 ms, inspiratory effort 1 and 2 cm H2O), a full-term infant (resistance 50 cm H2O/L/s, compliance 5 mL/cm H2O, TI 500 ms, inspiratory effort 2 and 4 cm H2O), and a child (resistance 30 cm H2O/L/s, compliance 10 mL/cm H2O, TI 600 ms, inspiratory effort 5 and 10 cm H2O). Two PS levels were tested (10 and 15 cm H2O) with and without leaks and with and without the leak compensation algorithm activated. RESULTS: Without leaks, only 2 neonatal ventilators and one adult ventilator had trigger delays under a given predefined acceptable limit (1/8 TI). Pressurization showed high variability between ventilators. Most ventilators showed TI in excess high enough to seriously impair patient-ventilator synchronization (> 50% of the TI of the subject). In some ventilators, leaks led to autotriggering and impairment of ventilation performance, but the influence of leaks was generally lower in neonatal ventilators. When a noninvasive ventilation algorithm was available, this was partially corrected. In general, tidal volume was calculated too low by the ventilators in the presence of leaks; the noninvasive ventilation algorithm was able to correct this difference in only 2 adult ventilators. CONCLUSIONS: No ventilator performed equally well under all tested conditions for all explored parameters. However, neonatal ventilators tended to perform better in the presence of leaks. These findings emphasize the need to improve algorithms for assisted ventilation modes to better deal with situations of high airway resistance, low pulmonary compliance, and the presence of leaks.

Adaptive intelligent agent approach to guide the web navigation on the PLAN-G distance learning platform

Hypermedia systems based on the Web for open distance education are becoming increasinglypopular as tools for user-driven access learning information. Adaptive hypermedia is a new direction in research within the area of user-adaptive systems, to increase its functionality by making it personalized [Eklu 961. This paper sketches a general agents architecture to include navigationaladaptability and user-friendly processes which would guide and accompany the student during hislher learning on the PLAN-G hypermedia system (New Generation Telematics Platform to Support Open and Distance Learning), with the aid of computer networks and specifically WWW technology [Marz 98-1] [Marz 98-2]. The PLAN-G actual prototype is successfully used with some informatics courses (the current version has no agents yet). The propased multi-agent system, contains two different types of adaptive autonomous software agents: Personal Digital Agents {Interface), to interacl directly with the student when necessary; and Information Agents (Intermediaries), to filtrate and discover information to learn and to adapt navigation space to a specific student

Altered adaptive but not veridical decision-making in substance dependent individuals.

Drug addiction is associated with impaired judgment in unstructured situations in which success depends on self-regulation of behavior according to internal goals (adaptive decision-making). However most executive measures are aimed at assessing decision-making in structured scenarios, in which success is determined by external criteria inherent to the situation (veridical decision-making). The aim of this study was to examine the performance of Substance Abusers (SA, n = 97) and Healthy Comparison participants (HC, n = 81) in two behavioral tasks that mimic the uncertainty inherent in real-life decision-making: the Cognitive Bias Task (CB) and the Iowa Gambling Task (IGT) (administered only to SA). A related goal was to study the interdependence between performances on both tasks. We conducted univariate analyses of variance (ANOVAs) to contrast the decision-making performance of both groups; and used correlation analyses to study the relationship between both tasks. SA showed a marked context-independent decision-making strategy on the CB's adaptive condition, but no differences were found on the veridical conditions in a subsample of SA (n = 34) and HC (n = 22). A high percentage of SA (75%) also showed impaired performance on the IGT. Both tasks were only correlated when no impaired participants were selected. Results indicate that SA show abnormal decision-making performance in unstructured situations, but not in veridical situations.