Ebook: Building Continents of Knowledge in Oceans of Data: The Future of Co-Created eHealth

This volume of Studies in Health Technology and Informatics – Building Continents of Knowledge in Oceans of Data: The Future of Co-Created eHealth – Proceedings of MIE 2018 – contributes to the discussion of ongoing challenges in eHealth, digitalization, capacity building and user engagement with true inter-disciplinary and cross-domain collaboration arising for 21-century health care.

The MIE conferences are the most important conference hosted by the European Federation for Medical Informatics (EFMI). The first MIE conference was hosted in Cambridge, UK in 1978 and therefore, MIE2018 marks the 40th anniversary for the Medical Informatics Europe – MIE – conferences. Over the years, the contributions presented at the MIE conferences, presented in the IOS Press series Studies in Health Technology and Informatics, has contributed to set the stage for medical informatics, health informatics and eHealth development in Europe.

The contributions in this book are no exception and share the full range of methodological and application oriented health informatics achievements at regional, national, and international level. The first part of the MIE 2018 conference theme – Building Continents of Knowledge in Oceans of Data – zoom in on experiences, methods and expectations for the future coming from achievements in Big Data analytics, data driven development of eHealth in the current health care systems and contributions from citizens' engagement in their self-management. The second part of the theme – The Future of Co-Created eHealth – has triggered a large number of papers describing strategies to create, evaluate, adjust or deliver tools and services for improvements in the healthcare organizations and citizens to respond to the challenges of the health systems.

This book presents the 190 full papers presented at that conference, held in Gothenburg, Sweden in April 2018. We have grouped the papers under these headings: Standards and interoperability, Implementation and evaluation, Knowledge management, Decision support, Modeling and analytics, Health informatics education and learning systems; and Patient centered services. Attention is also given to development for sustainable use, educational strategies and workforce development, which may arise when health professionals collaborate with colleagues and patients in virtual teams.

The concept of ‘meaningful use’ opens up additional perspectives and offers exciting opportunities to ensure that users – broadly understood as health providers, patients and their families or consumers at large – are offered workable solutions relevant to their needs.

This volume will be of interest to all those whose work involves the analysis and use of data to support more effective delivery of healthcare.

Paris/Stockholm/Örebro/Oslo, March 8th 2018

Adrien Ugon, ESIEE-Paris, Noisy-le-Grand, France

Daniel Karlsson, Linköping University, Linköping, Sweden

Gunnar O. Klein, Informatics/eHealth, School of Business, Örebro University

Anne Moen, Institute for Health and Society, University of Oslo, Norway

Using claims data for research is well established. However, most claims data analyses are focused on single countries. Multi-national approaches are scarce. The application of different anonymization techniques before data are shared for research as well as differences in the reimbursement systems hamper the use of claims data from multiple countries. This paper analyses data conflicts that occur when international claims data sets are used for research and develops a generic process to detect and resolve these conflicts. The approach was successfully applied in the EU-funded ADVOCATE (Added Value for Oral Care) project that acquired data from health insurance providers, health funds or health authorities in six European countries.

Malaria is an infectious disease affecting people across tropical countries. In order to devise efficient interventions, surveillance experts need to be able to answer increasingly complex queries integrating information coming from repositories distributed all over the globe. This, in turn, requires extraordinary coding abilities that cannot be expected from non-technical surveillance experts. In this paper, we present a deployment of Semantic Automated Discovery and Integration (SADI) Web services for the federation and querying of malaria data. More than 10 services were created to answer an example query requiring data coming from various sources. Our method assists surveillance experts in formulating their queries and gaining access to the answers they need.

Electronic exchange of medical data between clinics and test centers makes the testing process more efficient, enables continuity of care record and reuse of medical data. The presented project employs HL 7 FHIR approach to model clinical concepts for the medical data exchange between a test center and different hospitals. Using a standard FHIR editor we have modeled 1226 observation profiles, 2396 commercial tests profiles that are mapped to 3249 production tests profiles. We have also defined a concept of an order and developed RESTfull API protocol to facilitate the ordering process. Now the data exchange system is in production and processes more than 20 000 test orders with more than 40 000 tests a day.

Development of biobanks is still hampered by difficulty to collect high quality sample annotations using patient clinical information. The IBCB project evaluated the feasibility of a nationwide clinical data research network for this purpose. Method: the infrastructure, based on eHOP and I2B2 technologies, was interfaced with the legacy IT components of 3 hospitals. The evaluation focused on the data management process and tested 5 expert queries in Hepatocarcinoma. Results: the integration of biobank data was comprehensive and easy. Five out of 5 queries were successfully performed and shown consistent results with the data sources excepted one query which required to search in unstructured data. The platform was designed to be scalable and showed that with few effort biobank data and clinical data can be integrated and leveraged between hospitals. Clinical or phenotyping concepts extraction techniques from free text could significantly improve the samples annotation with fine granularity information.

Predictive models can support physicians to tailor interventions and treatments to their individual patients based on their predicted response and risk of disease and help in this way to put personalized medicine into practice. In allogeneic stem cell transplantation risk assessment is to be enhanced in order to respond to emerging viral infections and transplantation reactions. However, to develop predictive models it is necessary to harmonize and integrate high amounts of heterogeneous medical data that is stored in different health information systems. Driven by the demand for predictive instruments in allogeneic stem cell transplantation we present in this paper an ontology-based platform that supports data owners and model developers to share and harmonize their data for model development respecting data privacy.

We introduce 3000PA, a clinical document corpus composed of 3,000 EPRs from three different clinical sites, which will serve as the backbone of a national reference language resource for German clinical NLP. We outline its design principles, results from a medication annotation campaign and the evaluation of a first medication information extraction prototype using a subset of 3000PA.

Healthcare systems are costly in many countries, and hospitals have always been one of the major cornerstones of the healthcare industry. Medical supplies expense is an increasingly substantial category of hospital costs. In China, the expense of medical supplies is being controlled at the hospital level. Different from drug prescription, medical supplies utilization is not being standardized or guided by clinical guidelines. In order to achieve the goal, many hospitals directly disable the use of the most expensive medical supplies. One missing piece in consideration is the patient heterogeneity, which decides the medical necessity for a specific surgery/procedure and the associated medical supplies requirement. The other challenge is to justify the substitutability of the medical supplies being replaced. In this study, we explore a clinical similarity based framework to analyze the inpatient medical supplies use records and detect unnecessary utilization. More specifically, the inpatient stays are clustered based on patients' clinical conditions. After clustering, inpatient cases within each sub-group should have similar clinical necessities and therefore similar medical supplies utilization patterns. Thus the unnecessary medical utilization can be identified and the cost reduction suggestions can be provided accordingly. This framework will be illustrated though a case study of 3-year inpatient records from a Chinese hospital.

Collecting Patient Reported Outcomes (PROs) is generally seen as an effective way to assess the efficacy and appropriateness of medical interventions, from the patients' perspective. In 2016 the Galeazzi Orthopaedic Institute established a digitized program of PROs collection from spine, hip and knee surgery patients. In this work, we re-port the findings from the data analysis of the responses collected so far about the complementarity of PROs with respect to the data reported by the clinicians, and about the main biases that can undermine their validity and reliability. Although PROs collection is recognized as being far more complex than just asking the patients “how they feel” on a regular basis and it entails costs and devoted electronic platforms, we advocate their further diffusion for the assessment of health technology and clinical procedures.

In the Lille University Hospital (North of France), data from the Anesthesia Information Management System (Diane® are linked to the Hospital Information System and stored in a dedicated data warehouse since 2010. These electronic medical records need to be reused and analyzed for observational studies. The aim of this paper is to describe the framework developed to structure the operation of that anesthesia data warehouse for research purposes. The presented framework is structured around three meetings between clinicians, computer scientists, and statisticians. The data scientist acts as a coordinator, leads meetings, and checks each milestone. Reuse of anesthesia-related electronic medical record for research purposes is only allowed through this framework. The aim of the first meeting is to decide the primary and secondary objectives of the study. The aim of the second meeting is to validate the statistical protocol. The data are extracted and the statistical analyses are performed. Finally, the results are presented, explained and discussed during the third meeting. During a 6 months period, 27 projects were included in the framework leading to 5 scientific communications. As a result, case studies with extraction and/or analysis situations are presented. This collaboration led to an empowerment process between all three actors, which increased efficiency of the workflow. Implementation of this framework will keep encouraging collaborative publication in order to provide reproducible research evidence.

Epilepsy diagnosis is typically performed through 2Dvideo-EEG monitoring, relying on the viewer's subjective interpretation of the patient's movements of interest. Several attempts at quantifying seizure movements have been performed in the past using 2D marker-based approaches, which have several drawbacks for the clinical routine (e.g. occlusions, lack of precision, and discomfort for the patient). These drawbacks are overcome with a 3D markerless approach. Recently, we published the development of a single-bed 3Dvideo-EEG system using a single RGB-D camera (Kinect v1). In this contribution, we describe how we expanded the previous single-bed system to a multi-bed departmental one that has been managing 6.61 Terabytes per day since March 2016. Our unique dataset collected so far includes 2.13 Terabytes of multimedia data, corresponding to 278 3Dvideo-EEG seizures from 111 patients. To the best of the authors' knowledge, this system is unique and has the potential of being spread to multiple EMUs around the world for the benefit of a greater number of patients.

Adverse drug events (ADEs) are critical. Approximately 10% of fatal ADEs are believed to be allergic reactions. Therefore, sharing patient allergy information is beneficial to medical staff members in avoiding potentially lethal complications. We previously performed a nationwide study of patient allergy information in Japanese hospitals. The report showed that most of the responding hospitals needed a standard format for reporting the information. To establish this, we implemented a novel format for recording patient allergy information into the hospital information system at Tohoku University Hospital; this format was created through vigorous discussion among medical staff members with a variety of specialties, including doctors, nurses, pharmacists, nutritionists, and medical safety managers. In this study, we followed the amount of inputted allergy information and the number of incidents involving medication after implementation. The amount of allergy information inputted increased slightly. Although incidents involving medication also increased slightly, ADEs due to allergy significantly decreased. We believe that our findings will be useful in helping to determine the optimal characteristics of drug allergy information and to improve the dissemination of information regarding potential allergens and subsequent ADEs.

HAITooL information system design and implementation was based on Design Science Research Methodology, ensuring full participation, in close collaboration, of researchers and a multidisciplinary team of healthcare professionals. HAITooL enables effective monitoring of antibiotic resistance, antibiotic use and provides an antibiotic prescription decision-supporting system by clinicians, strengthening the patient safety procedures. The design, development and implementation process reveals benefits in organizational and behavior change with significant success. Leadership commitment multidisciplinary team and mainly informaticians engagement was crucial to the implementation process. Participants' motivation and the final product delivery and evolution depends on that.

Models of child primary health care vary across Europe. There are three categories, primary care paediatricians, general practitioner based, or mixed. This paper describes the metadata schema used in the profiling process of candidate data sources for appraisal for the Models of Child Health Appraised (MOCHA) project using the MOCHA International Research Opportunity Instrument (MIROI). The ten clinical indicators included: asthma, antibiotic stewardship, immunisation, rickets, diarrhea, epilepsy, depression, ADHD, enuresis and care of women during pregnancy. Our metadata allows us to identify data within included data sources concerning any of the 10 clinical indicators identified for comparative analysis within the MOCHA project. From the 30 countries we found a minimum of 5 and a maximum of 36 different databases for each indicator.

In this communication we identify strategies for effectively documenting Sexual Orientation and Gender Identity in Electronic Health Records. For this review a multidisciplinary group composed by three physicians, a nurse, an engineer and a lawyer analyzed the evidence in bibliography related to the topic and summarized the results. After analyzing the information, we summarized and classified them into three major topics: To request, to store and to display and access to the information. How to standardize those data and where data specifically will be populated in EHRs have not been answered yet. The target of all of these efforts should be: to be sensitive with the needs of the patient and to ensure high quality of care.

High accessibility of Electronic Health Record systems can increase usability but creates simultaneously patients' anxieties about privacy issues. In order to reduce the privacy concerns, we focused on control and awareness, and designed an approach that can provide availability of patient's clinical data to doctors in two scenarios; (S1) direct control by the patient when they are conscious, (S2) control by a trusted representative when the patient is unconscious. In this paper, we show further analysis in a survey (n = 310, age range: 19-91) done to test the acceptability of our concept of a using a trusted representative and to further understand the concerns of Japanese citizens to improve our system design. These results in S1 suggest that patients concerned about control have a stronger inclination to also choose full awareness. We found also that patients tended to choose the same level of awareness for the representative as they did for themselves in S2. In addition, patients who chose awareness in S1 tended to choose the same for their representative in S2 and themselves after recovery from unconsciousness. We also discuss the significant differences found between the age-groups 20-39 and 60-79. We conclude that the system design of privacy aware EHR systems must be improved to consider patients who want to preserve their choice of control in the event they become unconscious but do not want to use a representative to maintain control.

While there is a clear need to apply data analytics in the healthcare sector, this is often difficult because it requires combining sensitive data from multiple data sources. In this paper, we show how the cryptographic technique of secure multi-party computation can enable such data analytics by performing analytics without the need to share the underlying data. We discuss the issue of compliance to European privacy legislation; report on three pilots bringing these techniques closer to practice; and discuss the main challenges ahead to make fully privacy-preserving data analytics in the medical sector commonplace.

Introduction The new General Data Protection Regulation (GDPR) compels health care institutions and their software providers to properly document all personal data processing and provide clear evidence that their systems are inline with the GDPR. All applications involved in personal data processing should therefore produce meaningful event logs that can later be used for the effective auditing of complex processes. Aim This paper aims to describe and evaluate HS.Register, a system created to collect and securely manage at scale audit logs and data produced by a large number of systems. Methods HS.Register creates a single audit log by collecting and aggregating all kinds of meaningful event logs and data (e.g. ActiveDirectory, syslog, log4j, web server logs, REST, SOAP and HL7 messages). It also includes specially built dashboards for easy auditing and monitoring of complex processes, crossing different systems in an integrated way, as well as providing tools for helping on the auditing and on the diagnostics of difficult problems, using a simple web application. HS.Register is currently installed at five large Portuguese Hospitals and is composed of the following open-source components: HAproxy, RabbitMQ, Elasticsearch, Logstash and Kibana. Results HS.Register currently collects and analyses an average of 93 million events per week and it is being used to document and audit HL7 communications. Discussion Auditing tools like HS.Register are likely to become mandatory in the near future to allow for traceability and detailed auditing for GDPR compliance.

The Nordic eHealth Research Network, a subgroup of the Nordic Council of Ministers eHealth group, is working on developing indicators to monitor progress in availability, use and outcome of eHealth applications in the Nordic countries. This paper reports on the consecutive analysis of National eHealth policies in the Nordic countries from 2012 to 2016. Furthermore, it discusses the consequences for the development of indicators that can measure changes in the eHealth environment arising from the policies. The main change in policies is reflected in a shift towards more stakeholder involvement and intensified focus on clinical infrastructure. This change suggests developing indicators that can monitor understandability and usability of eHealth systems, and the use and utility of shared information infrastructure from the perspective of the end-users – citizens/patients and clinicians in particular.

The concerns about privacy and personal data protection resulted in reforms of the existing legislation in European Union (EU). The General Data Protection Regulation (GDPR) aims to reform the existing measures on the topic of personal data protection of the European Union citizens, with a strong input on the rights and freedoms of people and in the establishment of rules for the processing of personal data. OpenEHR is a standard that embodies many principles of interoperable and secure software for electronic health records. This work aims to understand to what extent the openEHR standard can be considered a solution for the requirements needed by GDPR. A list of requirements for a Hospital Information Systems (HIS) compliant with GDPR and an identification of openEHR specifications was made. The requirements were categorized and compared with the specifications. The requirements identified for the systems were matched with the openEHR specifications, which result in 16 requirements matched with openEHR. All the specifications identified matched at least one requirement. OpenEHR is a solution for the development of HIS that reinforce privacy and personal data protection, ensuring that they are contemplated in the system development. The institutions can secure that their Eletronic Health Record are compliant with GDPR while safeguarding the medical data quality and, as a result, the healthcare delivery.

Acute hospital admission among the elderly population is very common and have a high impact on the health services and the community, as well as on the individuals. Several studies have focused on the possible risk factors, however, predicting who is at risk for acute hospitalization associated with disease and symptoms is still an open research question. In this study, we investigate the use of machine learning algorithms for predicting acute admission in older people based on admission data from individual citizens 70 years and older who were hospitalized in the acute medical unit of Svendborg Hospital in Denmark.

Using gene markers and other patient features to predict clinical outcomes plays a vital role in enhancing clinical decision making and improving prognostic accuracy. This work uses a large set of colorectal cancer patient data to train predictive models using machine learning methods such as random forest, general linear model, and neural network for clinically relevant outcomes including disease free survival, survival, radio-chemotherapy response (RCT-R) and relapse. The most successful predictive models were created for dichotomous outcomes like relapse and RCT-R with accuracies of 0.71 and 0.70 on blinded test data respectively. The best prediction models regarding overall survival and disease-free survival had C-Index scores of 0.86 and 0.76 respectively. These models could be used in the future to aid a decision for or against chemotherapy and improve survival prognosis. We propose that future work should focus on creating reusable frameworks and infrastructure for training and delivering predictive models to physicians, so that they could be readily applied to other diseases in practice and be continuously developed integrating new data.

This paper describes a technology for predicting the aggravation of diabetic nephropathy from electronic health record (EHR). For the prediction, we used features extracted from event sequence of lab tests in EHR with a stacked convolutional autoencoder which can extract both local and global temporal information. The extracted features can be interpreted as similarities to a small number of typical sequences of lab tests, that may help us to understand the disease courses and to provide detailed health guidance. In our experiments on real-world EHRs, we confirmed that our approach performed better than baseline methods and that the extracted features were promising for understanding the disease.

Emergency room(ER) visit prediction, especially whether visit ER or not and ER visit count, is crucial for hospitals to reasonably adapt resource allocation and` for patients to know future health state. Some existing studies have explored to use machine learning methods especially kinds of general linear model to settle down the task. But, in the clinical problems, there exist complex correlation between targets and features. Generally, liner model is difficult to model complex correlation to make better prediction. Hence, in this paper, we propose to use two non-linear models to settle the problem, which are XGBoost and Recurrent Neural Network. Experimental results show both methods have better performance.

Patient experience is an emerging concept that supports the improvement of healthcare services through identified patient expectations and experiences. In addition to structured feedback through official channels, experiences about healthcare appear increasingly in digital services and social media. We explore a new patient experience harvesting process based on linguistic patterns to identify relevant expressions in online discussions about children's health. Our results from the analysis of 98 229 unique sentences suggests that the 7-step process can be useful in discovering patients' evaluations of their care experiences. We propose ways to extend the process to other care contexts by adjusting the semantic reference models.