Ebook: e-Health – For Continuity of Care
Information technology and the information sciences have been part of our lives for some time now. They have revolutionized the healthcare system, changing the whole health landscape, as well as health culture. New devices, sources of data and roles for all those involved in healthcare are being developed as a result.
This book presents the proceedings of the 25th European Medical Informatics Conference, held in Istanbul, Turkey in August/September 2014. The conference aims to present the most recent developments in biomedical informatics. The book is divided into 15 sections, which include: decision support systems and clinical practice guidelines; improved healthcare through informatics; data analysis; mobile health; technology and system evaluation; and text mining. The final two sections present posters from the conference.
The book will be of interest to all those in the healthcare sector, researchers and practitioners alike, who develop, evaluate or work with information technology.
On behalf of the European Federation for Medical Informatics and the Turkish Medical Informatics Association, we are proud and happy to invite you to the 25th European Medical Informatics Conference in Istanbul.
The 25th edition of our conference – this is an important anniversary!
Twenty-five is the age of youth, but also the beginning of maturity; it is a time of transition. What better place to celebrate this anniversary than in Istanbul, with its unique history and its fascinating mixture of past and present, of old and new, of religions and cultures.
Information technology and the information sciences have also reached the age of maturity. They are revolutionizing the healthcare system and deeply changing the whole health landscape, as well as health culture. Boundaries are becoming less clear. There is no less well-defined state than the ‘patient’ state. We are sometimes healthy and sometimes sick, but most of the time we are somewhere in between – not so healthy. We should all care about staying healthy. The boundary between health and ill-health is becoming harder to draw. A similar situation exists between prevention and care. While the healthcare system has always predominantly been involved in the management of disease and care, a new trend has appeared, with citizens increasingly involved in wellness and the prevention of ill-health. The quantified self provides the necessary tools to accompany this societal transformation.
The most recent impacts of the emerging field of biomedical informatics will be presented during the MIE 2014 conference. New phenotypical data sources, especially those arising from the ‘healthy world’, which is not usual for the medical sciences, are becoming an important pillar of health determinants, together with genetics, environment and lifestyle. New devices, new sources of data, new roles for all the actors involved are being invented. This point is well illustrated by several reports on Google Glass. The difference between Second Sight glasses, which were available as early as 2002, and Google Glass in 2014 is aesthetic, connectivity and information. The technology was already there, but only the added value derived from connectivity and information have made the glasses an attractive device. Environmental data, lifestyle, bio-captors, phenotypes extracted from EHRs, not to mention all the -omics, constitute the ‘holy grail’ of life science research: a data coverage of the major health determinants. Will we be able to transform the incredibly large amounts of data currently being made available into a clever and smart “Big data”. There are numerous challenges to be dealt with: technological challenges concerning the vast distributed volumes; semantic challenges concerning sources which are heterogeneous over time, of varying quality and reliability, and different contexts; ethical challenges when it comes to balancing public and individual benefits; and legal and societal challenges. All in all, the challenge of Big data questions the whole of society; its values, and its goals – not to mention the results. While this immature technology is promised a bright future, there are already enough reports stating that we need a more reliable methodology, especially when looking at preliminary work such as Google Flu Trends, which were demonstrated to be fast, inexpensive and wrong (Lazer et al., Science. 2014 Mar 14; 343(6176):1203–5. doi: 10.1126/science.1248506).
Care process reengineering, the involvement of citizens and patients, continuity of care and health: these are only some of the numerous revolutions taking place. This dynamic is characterized by combined advances in technology and sciences, deep socio-cultural changes, and wide global adoption. And this is precisely the heart of the challenge.
We do not have the option to be spectators; we are actors. Thus, it is necessary to move forward while building an evidence base; making sure that knowledge can be shared and that experiences are reproducible. Amassing evidence and using standards are thus important aspects of our work. It is necessary to keep the ultimate goal of improved healthcare in mind, all the time directing projects and activities towards this goal. The building up of evidence has to be done while retaining some room for disruptive ideas, non-acceptable hypotheses and non-demonstrable theories. It is also our challenge to allow room to all those who think differently, for those who don't respect the evidence, for those transgressing truth; because truth is a contextual reality, only valid in a certain time and context.
Finally, because health determinants are local, because most health threats are local, and because the process of globalisation tends to lose sight of the personal, we must also be sure to balance truly global science with local experience. To quote Ilias Iakovidis: “Medicine is a global science, and a local art”.
We would also like to acknowledge all our colleagues who devoted their time to the careful review of submissions. They have made it possible for the SPC to select the best papers and posters for inclusion in the proceedings. All submissions are Medline indexed and made available as open access through IOS Press homepage.
Istanbul, July 2014
Stig Kjær Andersen
The computerization of care pathways (CPs) has drawn considerable attention, for improving quality of health care and reducing costs. A well-known big challenge of implementing CPs is their flexibility and ad hoc variations in execution of clinical tasks. We observe that case management suits well to address this problem, and this paper proposes a CMMN-based CP model, where CMMN (Case Management Model and Notation) is becoming an industry standard. Via an experimental experience on modelling CHF (congestive heart failure) ambulatory CP, we illustrate that the usage of case management paves the way to popularize CPs, particularly for its quick deployment and execution in industrial products.
The modeling of clinical guidelines in order to apply them in computerized medical tools is a challenging and laborious task. In this project we show that conditional subordination links – a temporal relation concept of TimeML – can be used to describe condition-based activities in a guideline. Therefore, we extend the specification of TimeML concerning events and subordination links. Subsequently, linguistic and semantic rules are developed to automatically generate annotations for these links and classify them as relevant for the clinical care path. Finally, the evaluation of the method shows that this categorization supports the task of the guideline modeling expert.
Clinical practice guidelines (CPGs) are documents giving recommendations based on expert reasoning, weighing up the pros and cons of treatments on the basis of the available evidence. We propose a new approach to the construction of clinical decision support systems (CDSS), making use of the evidence-based medical reasoning used by experts in CPGs. In this study, we determined whether this approach could retrieve the recommendations for antibiotic prescription for empirical treatment in primary care. Methods: We manually extracted, from CPGs, all the properties of antibiotics underlying recommendations for their prescription or non-prescription. We then used these properties to establish an algorithm in the form of a sequence of conditions, leading to a list of recommended antibiotics. The optimal sequence was determined by studying, for each sequence, the degree of similarity between the list of antibiotics recommended in CPGs and the list obtained with the algorithm. Results: 12 antibiotic properties were used in the form of conditions in an algorithm. For 95% of clinical situations, 10 sequences retrieved the recommended treatment. Discussion: This algorithm could be used in a CDSS for antibiotic treatment and would be useful for experts drawing up CPGs.
Shared Decision Making (SDM) is the process of patients and clinicians working together to manage medical treatment using the existing knowledge base. This paper presents the YouCan framework, a system for summarizing and presenting the necessary knowledge for SDM related to pediatric cancer follow-up management. Knowledge modelling of a Clinical Practice Guideline produced a customized ontology, which was then passed through a pellet reasoned to produce a customized patient diary that summarizes a patient's oncological history as well as the potential issues they may face in follow-up.
Care pathways (CPs) as a means of healthcare quality control are getting increasing attention due to widespread recognition in the healthcare industry of the need for well coordinated, evidence based and personalized care. To keep the promise, CPs require continuous refinement in order to stay up to date with regard to both clinical guidelines and data-driven insights from real world practices. There is therefore a strong demand for a unified platform that allows harmonization of evidence coming from multiple sources. In this paper we describe Care Pathway Workbench, a web-based platform that enables users to build and continuously improve Case Management Model and Notation based CPs by harmonizing evidences from guidelines and patient data. To illustrate the functionalities, we describe how a CHF (Congestive Heart Failure) Ambulatory Care Pathway can be developed using this workbench by first extracting key elements from widely accepted guidelines for CHF management, then incorporating evidence mined from clinical practice data, and finally transforming and exporting the resulting CP model to a care management product.
Recently, National agencies in charge of the development of clinical practice guidelines (CPGs) have started to improve the usual narrative CPGs to provide guidance for different clinical pathways. In France, in conjunction with the development of the type 2 diabetes National CPGs, we have developed the system RecosDoc-Diabète which allows to interactively build a patient-centred pathway and get the appropriate recommendations. National narrative CPGs and RecosDoc-Diabète were published and made available online at the same time (February 2013). A questionnaire was provided to collect visitors' judgement about the system. Between February 12th and December 31st, 2013, 55,203 visitors accessed the narrative CPGs whereas 10,565 accessed the system. Among them, 186 (2%) responded to the questionnaire. One third of the comments were criticisms towards the CPG content. The system was globally positively evaluated although assessments were mixed illustrating that users' needs may be contradictory.
To enable the efficient reuse of standard based medical data we propose to develop a higher level information model that will complement the archetype model of ISO 13606. This model will make use of the relationships that are specified in UML to connect medical archetypes into a knowledge base within a repository. UML connectors were analyzed for their ability to be applied in the implementation of a higher level model that will establish relationships between archetypes. An information model was developed using XML Schema notation. The model allows linking different archetypes of one repository into a knowledge base. Presently it supports several relationships and will be advanced in future.
Every year, numerous clinical practice guidelines (CPGs) are published on a same topic. They may be conflicting, thus infringing clinicians' confidence in adhering to them. In order to build a clinical decision support system to assist GPs in the management of hypertension, we have considered three recent CPGs written in French. We developed a methodological framework to evaluate how consistent the three CPGs were. After a manual extraction of recommendation rules, all patient profiles covered by the CPGs have been identified. Then, ontological modeling and reasoning were used to build a subsumption graph of all profiles. This graph allows the retrieval of recommendations that could be conflicting. Results show that if rules are different in the three CPGs according to a document-based approach, many profiles are related through subsumption, and no critical inconsistencies were discovered when implementing an ontological modeling.
Evidence-based medical practice requires that clinical guidelines need to be documented in such a way that they represent a clinical workflow in its most accessible form. In order to optimize clinical processes to improve clinical outcomes, we propose a Service Oriented Architecture (SOA) based approach for implementing clinical guidelines that can be accessed from an Electronic Health Record (EHR) application with a Web Services enabled communication mechanism with the Enterprise Service Bus. We have used Business Process Modelling Notation (BPMN) for modelling and presenting the clinical pathway in the form of a workflow. The aim of this study is to produce spontaneous alerts in the healthcare workflow in the diagnosis of Chronic Obstructive Pulmonary Disease (COPD). The use of BPMN as a tool to automate clinical guidelines has not been previously employed for providing Clinical Decision Support (CDS).
Statistical hypothesis testing is an essential component of biological and medical studies for making inferences and estimations from the collected data in the study; however, the misuse of statistical tests is widely common. In order to prevent possible errors in convenient statistical test selection, it is currently possible to consult available test selection algorithms developed for various purposes. However, the lack of an algorithm presenting the most common statistical tests used in biomedical research in a single flowchart causes several problems such as shifting users among the algorithms, poor decision support in test selection and lack of satisfaction of potential users. Herein, we demonstrated a unified flowchart; covers mostly used statistical tests in biomedical domain, to provide decision aid to non-statistician users while choosing the appropriate statistical test for testing their hypothesis. We also discuss some of the findings while we are integrating the flowcharts into each other to develop a single but more comprehensive decision algorithm.
Many prognostic models are not adopted in clinical practice regardless of their reported accuracy. Doubts about the basis of the model is considered to be a major reason for this as the evidence behind clinical models is often not clear to anyone other than their developers. We propose a framework for representing the evidence behind Bayesian networks (BN) developed for prognostic decision support. The aim of this evidence framework is to be able to present all the evidence alongside the BN itself. We illustrate this framework by a BN developed with clinical evidence to predict coagulation disorders in trauma care.
The complexity of the medical diagnosis is faced by practitioners relying mainly on their experiences. This can be acquired during daily practices and on-the-job training. Given the complexity and extensiveness of the subject, supporting tools that include knowledge extracted by highly specialized practitioners can be valuable. In the present work, a Decision Support System (DSS) for hand dermatology was developed based on data coming from a Visit Report Form (VRF). Using a Bayesian approach and factors significance difference over the population average for the case, we demonstrated the potentiality of creating an enhanced VRF that include a diagnoses distribution probability based on the DSS rules applied for the specific patient situation.
Medical decision making, such as choosing which drugs to prescribe, requires to consider mandatory constraints, e.g. absolute contraindications, but also preferences that may not be satisfiable, e.g. guideline recommendations or patient preferences. The major problem is that these preferences are complex, numerous and come from various sources. The considered criteria are often conflicting and the number of decisions is too large to be explicitly handled. In this paper, we propose a framework for encoding medical preferences using a new connective, called ordered disjunction symbolized by ~×. Intuitively, the preference “Diuretic~×Betablocker means: “Prescribe a Diuretic if possible, but if this is not possible, then prescribe a Betablocker”. We give an inference method for reasoning about the preferences and we show how this framework can be applied to a part of a guideline for hypertension.
In Denmark, there are plans for establishing a national decision support system, providing on-line support for physicians during drug prescribing. This includes establishment of a national database containing information about each patient's drug allergies. Allergy information already exists in medication modules in hospital systems and primary care systems and thus constitutes a potential source for the national allergy database. This paper reports an analysis of local data structure, content and registration policies with the aim to re-use existing allergy data. The result of the analysis is that due to lack of harmonisation most existing cannot be re-used in the national database. The paper propose a common dataset for allergy data where national and international standards were considered.
Missed, wrong or delayed diagnosis has a direct effect on patient safety. Diagnostic errors have been discussed at length, however it still lacks a systematic approach. This study proposed a more systematic way of studying diagnostic errors by using a causal loop diagram. A systematic review was used to find the key factors which may cause diagnostic errors and their interrelationships. A causal loop diagram, as a qualitative model at the first stage of system dynamics modeling, was produced to map all the factor and interrelationships. The diagram provides not only the direct and indirect factors affecting correct diagnosis, but also a clear view of how the change of one factor in the model triggers changes of other factors and then the change of the number of final diagnostic errors.
The Radiology Gamuts Ontology (RGO) is a knowledge model of diseases, interventions, and imaging manifestations. RGO incorporates 16,822 terms with their synonyms and abbreviations and 55,393 relationships between terms. Subsumption defines the relationship between more general and more specific terms; causality relates disorders and their imaging manifestations. We explored the application of the RGO to build an interactive decision support system for radiological diagnosis. The Gamuts DDx system was created to apply the RGO's knowledge: it identifies a list of potential diagnoses in response to one or more user-specified imaging observations. The system also identifies a set of observations that allow one to narrow the diagnosis, and dynamically narrows or expands the list of diagnoses as imaging findings are selected or deselected. The functionality has been implemented as a web-based user interface and as a web service. The current work demonstrates the feasibility of exploiting the RGO's causal knowledge to provide interactive decision support for diagnosis of imaging findings. Ongoing efforts include the further development of the system's knowledge base and evaluation of the system in clinical use.
A personal health system platform for the management of patients with chronic liver disease that incorporates a novel approach to integrate decision support and guidance through care pathways for patients and their doctors is presented in this paper. The personal health system incorporates an integrated decision support engine that guides patients and doctors through the management of the disease by issuing tasks and providing recommendations to both the care team and the patient and by controlling the execution of a Care Flow Plan based on the results of tasks and the monitored health status of the patient. This Care Flow Plan represents a formal, business process based model of disease management designed off-line by domain experts on the basis of clinical guidelines, knowledge of care pathways and an organisational model for integrated, patient-centred care. In this way, remote monitoring and treatment are dynamically adapted to the patient's actual condition and clinical symptoms and allow flexible delivery of care with close integration of specialists, therapists and care-givers.
Neuro-fuzzy system is a combination of neural network and fuzzy system in such a way that neural network learning algorithms, is used to determine parameters of the fuzzy system. This paper describes the application of multiple adaptive neuro-fuzzy inference system (MANFIS) model which has hybrid learning algorithm for classification of hemiplegic gait acceleration (HGA) signals. Decision making was performed in two stages: feature extraction using the wavelet transforms (WT) and the ANFIS trained with the backpropagation gradient descent method in combination with the least squares method. The performance of the ANFIS model was evaluated in terms of training performance and classification accuracies and the results confirmed that the proposed ANFIS model has potential in classifying the HGA signals.
Clinical decision support systems are an important aspect of medical informatics. The increasing amount of available patient data requires physicians to rely on information technology for research and during their day by day work. In intensive care medicine, fast actions are especially important. One major step towards enabling direct interaction of medical staff with patient data was the development of clinical data repositories with easy query frontends. While clinical data repositories can be extended for the use of real-time data, the corresponding query frontends do not support the time concepts necessary for real-time queries and decision support. Aim of this project is the development of a user interface to give physicians visual understanding of propositional logic combined with time concepts. Thus, physicians should be able formulate simple time based queries on their own – and validate and quality check complex queries created by medical informatics experts.
In this paper we present our experiences with extending an existing approach for an archetype-compliant collection and export of data according to the openEHR specifications within the open source EHR system OpenMRS. It allows an automatic generation of forms from templates, which were introduced by openEHR as an extension of the dual-model approach. Data entered in these forms can be exported in form of standardized EHR extracts. The use of templates allowed us to solve problems reported for the original archetype-based version of the approach, which were caused by the high optionality within archetypes.
Aim: To describe the requirements, development and evaluation of a cognitive disorders and older persons' clinical and research application, outlining the conceptual and practical challenges. Methods: A technology development methodology was used to develop a database of people being investigated for or diagnosed with cognitive disorders as well as their carers. The methodology involved phases of requirements gathering, modeling and prototype creation, and ‘bench testing’ the prototype with experts. Results: This case study suggests that construction and population of a memory clinic and research database is feasible, but initial development is complex. Its utility can be evaluated to some extent and was found to be acceptable to most users. Discussion and Conclusions: The development of a system needs to take in account existing data collection methods and other information systems used. The GreyMatters system can be considered a supplementary or complementary health record that sits alongside the main Trust information system. Integrating data from multiple systems enhances utility to clinical and research users.
Adverse drug events (ADEs) are common, costly and one of the most important issues in contemporary pharmacotherapy. Current drug safety surveillance methods are largely based on spontaneous reports. However, this is known to be rather ineffective. There is a lack of automated systems checking potential ADEs on routine data captured in electronic health records (EHRs); present systems are usually built directly on top of specific clinical information systems through proprietary interfaces. In the context of the European project “SALUS”, we aim to provide an infrastructure as well as a tool-set for accessing and analyzing clinical patient data of heterogeneous clinical information systems utilizing standard methods. This paper focuses on two components of the SALUS architecture: The “Semantic Interoperability Layer” (SIL) enables an access to disparate EHR sources in order to provide the patient data in a common data model for ADE detection within the “ADE Detection and Notification Tool” (ANT). The SIL in combination with the ANT can be used in different clinical environments to increase ADE detection and reporting rates. Thus, our approach promises a profound impact in the domain of pharmacovigilance.