Search

Introduction: Blood cultures are often performed in the intensive care unit (ICU) to detect bloodstream infections and identify pathogen type, further guiding treatment. Early detection is essential, as a bloodstream infection can give cause to sepsis, a severe immune response associated with an increased risk of organ failure and death., Problem Statement: The early clinical detection of a bloodstream infection is challenging but rapid targeted treatment, within the first place antimicrobials, substantially increases survival chances. As blood cultures require time to incubate, early clinical detection using physiological signals combined with indicative lab values is pivotal., Objective: In this work, a novel method is constructed and explored for the potential prediction of the outcome of a blood culture test. The approach is based on a temporal computational model which uses nine clinical parameters measured over time., Methodology: We use a bidirectional long short-term memory neural network, a type of recurrent neural network well suited for tasks where the time lag between a predictive event and outcome is unknown. Evaluation is performed using a novel high-quality database consisting of 2177 ICU admissions at the Ghent University Hospital located in Belgium., Results: The network achieves, on average, an area under the receiver operating characteristic curve of 0.99 and an area under the precision-recall curve of 0.82. In addition, our results show that predicting several hours upfront is possible with only a small decrease in predictive power. In this setting, it outperforms traditional non-temporal, machine learning models., Conclusion: Our proposed computational model accurately predicts the outcome of blood culture tests using nine clinical parameters. Moreover, it can be used in the ICU as an early warning system to detect patients at risk of blood stream infection., (Copyright © 2018 Elsevier B.V. All rights reserved.)

Predicting the bed occupancy of an intensive care unit (ICU) is a daunting task. The uncertainty associated with the prognosis of critically ill patients and the random arrival of new patients can lead to capacity problems and the need for reactive measures. In this paper, we work towards a predictive model based on Random Survival Forests which can assist physicians in estimating the bed occupancy. As input data, we make use of the Sequential Organ Failure Assessment (SOFA) score collected and calculated from 4098 patients at two ICU units of Ghent University Hospital over a time period of four years. We compare the performance of our system with a baseline performance and a standard Random Forest regression approach. Our results indicate that Random Survival Forests can effectively be used to assist in the occupancy prediction problem. Furthermore, we show that a group based approach, such as Random Survival Forests, performs better compared to a setting in which the length of stay of a patient is individually assessed.

Introduction: This article is part of the Focus Theme of METHODS of Information in Medicine on "Managing Interoperability and Complexity in Health Systems"., Objectives: Handheld computers, such as tablets and smartphones, are becoming more and more accessible in the clinical care setting and in Intensive Care Units (ICUs). By making the most useful and appropriate data available on multiple devices and facilitate the switching between those devices, staff members can efficiently integrate them in their workflow, allowing for faster and more accurate decisions. This paper addresses the design of a platform for the efficient switching between multiple devices in the ICU. The key functionalities of the platform are the integration of the platform into the workflow of the medical staff and providing tailored and dynamic information at the point of care., Methods: The platform is designed based on a 3-tier architecture with a focus on extensibility, scalability and an optimal user experience. After identification to a device using Near Field Communication (NFC), the appropriate medical information will be shown on the selected device. The visualization of the data is adapted to the type of the device. A web-centric approach was used to enable extensibility and portability., Results: A prototype of the platform was thoroughly evaluated. The scalability, performance and user experience were evaluated. Performance tests show that the response time of the system scales linearly with the amount of data. Measurements with up to 20 devices have shown no performance loss due to the concurrent use of multiple devices., Conclusions: The platform provides a scalable and responsive solution to enable the efficient switching between multiple devices. Due to the web-centric approach new devices can easily be integrated. The performance and scalability of the platform have been evaluated and it was shown that the response time and scalability of the platform was within an acceptable range.

Background: An electronic decision support programme was developed within the intensive care unit (ICU) that provides an overview of all infection-related patient data, and allows ICU physicians to add clinical information during patient rounds, resulting in prospective compilation of a database., Aim: To assess the validity of computer-assisted surveillance (CAS) of ICU-acquired infection performed by analysis of this database., Methods: CAS was compared with prospective paper-based surveillance (PBS) for ICU-acquired respiratory tract infection (RTI), bloodstream infection (BSI) and urinary tract infection (UTI) over four months at a 36-bed medical and surgical ICU. An independent panel reviewed the data in the case of discrepancy between CAS and PBS., Findings: PBS identified 89 ICU-acquired infections (13 BSI, 18 UTI, 58 RTI) and CAS identified 90 ICU-acquired infections (14 BSI, 17 UTI, 59 RTI) in 876 ICU admissions. There was agreement between CAS and PBS on 13 BSI (100 %), 14 UTI (77.8 %) and 42 RTI (72.4 %). Overall, there was agreement on 69 infections (77.5%), resulting in a kappa score of 0.74. Discrepancy between PBS and CAS was the result of capture error in 11 and 14 infections, respectively. Interobserver disagreement on probability (13 RTI) and focus (two RTI, one UTI) occurred for 16 episodes. The time required to collect information using CAS is less than 30% of the time required when using PBS., Conclusion: CAS for ICU-acquired infection by analysis of a database built through daily workflow is a feasible surveillance method and has good agreement with PBS. Discrepancy between CAS and PBS is largely due to interobserver variability., (Copyright © 2014 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.)

Background: As the amount of information in electronic health care systems increases, data operations get more complicated and time-consuming. Intensive Care platforms require a timely processing of data retrievals to guarantee the continuous display of recent data of patients. Physicians and nurses rely on this data for their decision making. Manual optimization of query executions has become difficult to handle due to the increased amount of queries across multiple sources. Hence, a more automated management is necessary to increase the performance of database queries. The autonomic computing paradigm promises an approach in which the system adapts itself and acts as self-managing entity, thereby limiting human interventions and taking actions. Despite the usage of autonomic control loops in network and software systems, this approach has not been applied so far for health information systems., Methods: We extend the COSARA architecture, an infection surveillance and antibiotic management service platform for the Intensive Care Unit (ICU), with self-managed components to increase the performance of data retrievals. We used real-life ICU COSARA queries to analyse slow performance and measure the impact of optimizations. Each day more than 2 million COSARA queries are executed. Three control loops, which monitor the executions and take action, have been proposed: reactive, deliberative and reflective control loops. We focus on improvements of the execution time of microbiology queries directly related to the visual displays of patients' data on the bedside screens., Results: The results show that autonomic control loops are beneficial for the optimizations in the data executions in the ICU. The application of reactive control loop results in a reduction of 8.61% of the average execution time of microbiology results. The combined application of the reactive and deliberative control loop results in an average query time reduction of 10.92% and the combination of reactive, deliberative and reflective control loops provides a reduction of 13.04%., Conclusions: We found that by controlled reduction of queries' executions the performance for the end-user can be improved. The implementation of autonomic control loops in an existing health platform, COSARA, has a positive effect on the timely data visualization for the physician and nurse.

The Intensive Care Unit is a data intensive environment where large volumes of patient monitoring and observational data are daily generated. Today, there is a lack of an integrated clinical platform for automated decision support and analysis. Despite the potential of electronic records for infection surveillance and antibiotic management, different parts of the clinical data are stored across databases in their own formats with specific parameters, making access to all data a complex and time-consuming challenge. Moreover, the motivation behind physicians' therapy decisions is currently not captured in existing information systems. The COSARA research project offers automated data integration and services for infection control and antibiotic management for Ghent University Hospital. The platform not only gathers and integrates all relevant data, it also presents the information visually at the point of care. In this paper, we describe the design and value of COSARA for clinical treatment and infectious diseases monitoring. On the one hand, this platform can facilitate daily bedside follow-up of infections, antibiotic therapies and clinical decisions for the individual patient, while on the other hand, the platform serves as management view for infection surveillance and care quality improvement within the complete ICU ward. It is shown that COSARA is valuable for registration, real-time presentation and management of infection-related and antibiotics data.

As the complexity and amount of medical information keeps increasing, it is difficult to maintain the same quality of care. Therefore, clinical guidelines are used to structure best practices and care, but they also support physicians and nurses in the diagnostic and treatment process. Currently, no standardized format exists to represent these guidelines. Translating guidelines into a computer interpretable format can overcome problems in the physicians' workflow and improve clinician's uptake. An engine is proposed to automatically translate and execute clinical guidelines. These guidelines are represented as flowcharts, expressed in either (i) a computer interpretable guideline format or (ii) a UML diagram. A detailed overview of the architecture is presented and algorithms, aiming at grouping several components and distributing the guidelines, are proposed to optimize the execution of the guidelines. The Modified Schofield guideline for the calculation of the calorie need for burn patients was used for evaluation. Results show that the execution of guidelines using the engine is very efficient. Using optimization algorithms the execution times can be lowered., (Copyright © 2012 Elsevier Ltd. All rights reserved.)

The computerization of Intensive Care Units provides an overwhelming amount of electronic data for both medical and financial analysis. However, the current tarification, which is the process to tick and count patients' procedures, is still a repetitive, time-consuming process on paper. Nurses and secretaries keep track manually of the patients' medical procedures. This paper describes the design methodology and implementation of automated tarification services. In this study we investigate if the tarification can be modeled in service oriented architecture as a composition of interacting services. Services are responsible for data collection, automatic assignment of records to physicians and application of rules. Performance is evaluated in terms of execution time, cost evaluation and return on investment based on tracking of real procedures. The services provide high flexibility in terms of maintenance, integration and rules support. It is shown that services offer a more accurate, less time-consuming and cost-effective tarification.

The Intensive Care Unit (ICU) is an extremely data-intensive environment where each patient needs to be monitored 24/7. Bedside monitors continuously register vital patient values (such as serum creatinine, systolic blood pressure) which are recorded frequently in the hospital database (e.g. every 2 min in the ICU of the Ghent University Hospital), laboratories generate hundreds of results of blood and urine samples, and nurses measure blood pressure and temperature up to 4 times an hour. The processing of such large amount of data requires an automated system to support the physicians' daily work. The Intensive Care Service Platform (ICSP) offers the needed support through the development of medical support services for processing and monitoring patients' data. With an increased deployment of these medical support services, reusing existing services as building blocks to create new services offers flexibility to the developer and accelerates the design process. This paper presents a new addition to the ICSP, the Dynamic Composer for Web services. Based on a semantic description of the medical support services, this Composer enables a service to be executed by creating a composition of medical services that provide the needed calculations. The composition is achieved using various algorithms satisfying certain quality of service (QoS) constraints and requirements. In addition to the automatic composition the paper also proposes a recovery mechanism in case of unavailable services. When executing the composition of medical services, unavailable services are dynamically replaced by equivalent services or a new composition achieving the same result. The presented platform and QoS algorithms are put through extensive performance and scalability tests for typical ICU scenarios, in which basic medical services are composed to a complex patient monitoring service., (Copyright © 2010 Elsevier Ireland Ltd. All rights reserved.)

Background: Information technology (IT) may improve the quality, safety and efficiency of medicine, and is especially useful in intensive Care Units (ICUs) as these are extremely data-rich environments with round-the-clock changing parameters. However, data regarding the implementation rates of IT in ICUs are scarce, and restricted to non-European countries. The current paper aims to provide relevant information regarding implementation of IT in Flemish ICU's (Flanders, Belgium)., Methods: The current study is based on two separate but complementary surveys conducted in the region of Flanders (Belgium): a written questionnaire in 2005 followed by a telephone survey in October 2008. We have evaluated the actual health IT adoption rate, as well as its evolution over a 3-year time frame. In addition, we documented the main benefits and obstacles for taking the decision to implement an Intensive Care Information System (ICIS)., Results: Currently, the computerized display of laboratory and radiology results is almost omnipresent in Flemish ICUs, (100% and 93.5%, respectively), but the computerized physician order entry (CPOE) of these examinations is rarely used. Sixty-five % of Flemish ICUs use an electronic patient record, 41.3% use CPOE for medication prescriptions, and 27% use computerized medication administration recording. The implementation rate of a dedicated ICIS has doubled over the last 3 years from 9.3% to 19%, and another 31.7% have plans to implement an ICIS within the next 3 years. Half of the tertiary non-academic hospitals and all university hospitals have implemented an ICIS, general hospitals are lagging behind with 8% implementation, however. The main reasons for postponing ICIS implementation are: (i) the substantial initial investment costs, (ii) integration problems with the hospital information system, (iii) concerns about user-friendly interfaces, (iv) the need for dedicated personnel and (v) the questionable cost-benefit ratio., Conclusions: Most ICUs in Flanders use hospital IT systems such as computerized laboratory and radiology displays. The adoption rate of ICISs has doubled over the last 3 years but is still surprisingly low, especially in general hospitals. The major reason for not implementing an ICIS is the substantial financial cost, together with the lack of arguments to ensure the cost/benefit.

Background: Computerized ICUs rely on software services to convey the medical condition of their patients as well as assisting the staff in taking treatment decisions. Such services are useful for following clinical guidelines quickly and accurately. However, the development of services is often time-consuming and error-prone. Consequently, many care-related activities are still conducted based on manually constructed guidelines. These are often ambiguous, which leads to unnecessary variations in treatments and costs.The goal of this paper is to present a semi-automatic verification and translation framework capable of turning manually constructed diagrams into ready-to-use programs. This framework combines the strengths of the manual and service-oriented approaches while decreasing their disadvantages. The aim is to close the gap in communication between the IT and the medical domain. This leads to a less time-consuming and error-prone development phase and a shorter clinical evaluation phase., Methods: A framework is proposed that semi-automatically translates a clinical guideline, expressed as an XML-based flow chart, into a Drools Rule Flow by employing semantic technologies such as ontologies and SWRL. An overview of the architecture is given and all the technology choices are thoroughly motivated. Finally, it is shown how this framework can be integrated into a service-oriented architecture (SOA)., Results: The applicability of the Drools Rule language to express clinical guidelines is evaluated by translating an example guideline, namely the sedation protocol used for the anaesthetization of patients, to a Drools Rule Flow and executing and deploying this Rule-based application as a part of a SOA. The results show that the performance of Drools is comparable to other technologies such as Web Services and increases with the number of decision nodes present in the Rule Flow. Most delays are introduced by loading the Rule Flows., Conclusions: The framework is an effective solution for computerizing clinical guidelines as it allows for quick development, evaluation and human-readable visualization of the Rules and has a good performance. By monitoring the parameters of the patient to automatically detect exceptional situations and problems and by notifying the medical staff of tasks that need to be performed, the computerized sedation guideline improves the execution of the guideline.

The increasing complexity of procedures in the intensive care unit (ICU) requires complex software services, to reduce improper use of antibiotics and inappropriate therapies, and to offer earlier and more accurate detection of infections and antibiotic resistance. We investigated whether web-based software can facilitate the computerization of complex medical processes in the ICU. The COSARA application contains the following modules: Infection overview, Thorax, Microbiology, Antibiotic therapy overview, Admission cause with comorbidity and admission diagnosis, Infection linking and registration, and Feedback. After the implementation and test phase, the COSARA software was installed on a physician's office PC and then on the bedside PCs of the patients. Initial evaluation indicated that the services had been integrated easily into the daily clinical workflow of the medical staff. The use of a service oriented architecture with web service technology for the development of advanced decision support in the ICU offers several advantages over classical software design approaches.

The importance of computer aided decision making is continuously increasing. In the ICU, medical decision support services gather and process medical data of patients and present results and suggestions to the medical staff. The medical decision support services can monitor for example blood pressure, creatinine levels or the usage of antibiotics. If certain levels are crossed, they raise alerts so that the medical staff can take appropriate actions if required. This significantly reduces the amount of data needing to be processed by the medical staff. To handle the large amount of data that is generated by the ICU on a daily basis, a platform for routing and processing this data is necessary. In this paper we propose a platform based on JAIN SLEE and an Enterprise Service Bus. The platform takes care of the routing of the data to the appropriate services and allows to easily deploy and manage services. In this paper, we present the design details and the evaluation results. Furthermore, it is shown that the platform is capable of routing and processing all the events generated by the ICU within strict time constraints.

Objectives: This paper addresses the design of a platform for the management of medical decision data in the ICU. Whenever new medical data from laboratories or monitors is available or at fixed times, the appropriate medical support services are activated and generate a medical alert or suggestion to the bedside terminal, the physician's PDA, smart phone or mailbox. Since future ICU systems will rely ever more on medical decision support, a generic and flexible subscription platform is of high importance., Methods: Our platform is designed based on the principles of service-oriented architectures, and is fundamental for service deployment since the medical support services only need to implement their algorithm and can rely on the platform for general functionalities. A secure communication and execution environment are also provided., Results: A prototype, where medical support services can be easily plugged in, has been implemented using Web service technology and is currently being evaluated by the Department of Intensive Care of the Ghent University Hospital. To illustrate the platform operation and performance, two prototype medical support services are used, showing that the extra response time introduced by the platform is less than 150 ms., Conclusions: The platform allows for easy integration with hospital information systems. The platform is generic and offers user-friendly patient/service subscription, transparent data and service resource management and priority-based filtering of messages. The performance has been evaluated and it was shown that the response time of platform components is negligible compared to the execution time of the medical support services.

This paper addresses the design of a generic and scalable platform for the execution of medical decision support agents in the intensive care unit (ICU). As will be motivated, medical decision support agents can impose high computational load and in practical setups a large amount of such agents are typically running in parallel. Future ICU systems will rely on extensive medical decision support. However, in current systems only one workstation is typically dedicated for the execution of medical decision support agents. Therefore, we propose an architecture based on middleware technology to allow for easy distribution of the agents along multiple workstations. The architecture allows for easy integration with a general ICU data flow management architecture.

Acute kidney injury (AKI) is very common among critically-ill patients and is correlated with significant morbidity and mortality. The RIFLE criteria (an acronym comprising Risk, Injury, Failure, Loss and End-stage kidney disease), were developed by a panel of experts aiming at standardizing the definition of AKI and to subdivide AKI into different categories of severity. However, although these criteria are clear and easy to understand, they are still complex and labour-intensive, and therefore mostly used in retrospective. The use of an electronic alert based on the RIFLE criteria, which warns the physician in real-time when kidney function is deteriorating can help to implement these criteria in daily clinical practice. In this paper we describe the successful implementation of such an alert system. Not only were there technological barriers to solve; also acceptance of the alert by the end user was of pivotal importance. Further research is currently performed to investigate whether the implementation of real-time electronic RIFLE alerts induce faster therapeutic intervention, and to evaluate the impact of a more timely intervention on improved preservation of kidney function and patients' outcome.

Objectives: The current Intensive Care Information Systems (IC-ISs) collect and store monitoring data in on automated way and can replace all paper forms by an electronic equivalent, resulting in a paperless ICU. Future development of IC-ISs will now have to focus on bedside clinical decision support. The current IC-ISs are data-driven systems, with a two-layer software architecture. This software architecture is hardly maintainable and probably not the most optimal architecture to make the transition towards future systems with-decision support. The aim of this research was to address the design of an alternative software architecture based on new paradigms., Methods: State-of-the art component, middleware and agent technology were deployed to design and implement a software architecture for ICU data flow management., Results: An advanced multi-layer architecture for efficient data flow management in the ICU has been designed. The architecture is both generic and scalable, which means that it neither depends on a particular ICU nor on the deployed monitoring devices. Automatic device detection and Graphical User Interface generation are taken into account. Furthermore, a demonstrator has been developed as a proof that the proposed conceptual software architecture is feasible in practice. The core of the new architecture consists of Bed Decision Agents (BDAs). The introduction of BDAs, who perform specific dedicated tasks, improves the adaptability and maintainability of the future very complex IC-ISs., Conclusions: A software architecture, based on component, middleware and agent technology, is feasible and offers important advantages over the currently used two-layer software architecture.

Background: Echo-state networks (ESN) are part of a group of reservoir computing methods and are basically a form of recurrent artificial neural networks (ANN). These methods can perform classification tasks on time series data. The recurrent ANN of an echo-state network has an 'echo-state' characteristic. This 'echo-state' functions as a fading memory: samples that have been introduced into the network in a further past, are faded away. The echo-state approach for the training of recurrent neural networks was first described by Jaeger H. et al. In clinical medicine, until this moment, no original research articles have been published to examine the use of echo-state networks. Methods: This study examines the possibility of using an echo-state network for prediction of dialysis in the ICU. Therefore, diuresis values and creatinine levels of the first three days after ICU admission were collected from 830 patients admitted to the intensive care unit (ICU) between May 31th 2003 and November 17th 2007. The outcome parameter was the performance by the echo-state network in predicting the need for dialysis between day 5 and day 10 of ICU admission. Patients with an ICU length of stay <10 days or patients that received dialysis in the first five days of ICU admission were excluded. Performance by the echo-state network was then compared by means of the area under the receiver operating characteristic curve (AUC) with results obtained by two other time series analysis methods by means of a support vector machine (SVM) and a naive Bayes algorithm (NB). Results: The AUC's in the three developed echo-state networks were 0.822, 0.818, and 0.817. These results were comparable to the results obtained by the SVM and the NB algorithm. Conclusions: This proof of concept study is the first to evaluate the performance of echo-state networks in an ICU environment. This echo-state network predicted the need for dialysis in ICU patients. The AUC's of the echo-state networks were good and comparable to the performance of other classification algorithms. Moreover, the echo-state network was more easily configured than other time series modeling technologies. [ABSTRACT FROM AUTHOR]