Ebook: Health Informatics Meets eHealth

Digital Insight – Information-Driven Health & Care

A hospital is generally regarded as a place where patients are cured from injuries and diseases. However, in an increasing number of cases, clinical patient management within the hospital alone is not enough for successful treatment, but continuation of the therapy beyond the hospital walls is indicated. This requires not only cooperative and motivated patients, but also information exchange with out-patient healthcare providers and professionals. Apart from physicians, also nurses are concerned, since nurses usually have more contact to the patients and their families. Today's ICT tools need to actively involve all the players in healthcare, in order to close the gaps in between multimodal patient care. Additionally, data should be complemented with data from the patients' homes and support active and independent living.

Ineffective discharge management may jeopardize successful conclusion of treatment. Moreover, hospital stays can represent independent risk factors for additional diseases or complications, e.g. when nosocomial infections or delirium are involved. Many of these events could potentially be prevented, but prevention strategies often need an integrated view and involve more than just one type of hospital personnel, i.e. physicians, nurses, administration, technicians, etc. Within hospitals, a lot of data that could help to early detect or even to prevent such events are already available. So far, however, often these data are either not sufficiently considered or implications of identified risk factors are not sufficiently communicated. “Information-driven Health & Care” relates both, a) identification of certain factors within data pools from health & care and b) information exchange among all the involved stakeholders. ICT can support both aspects and provide tools that help health professions in identifying and communicating relevant data. Such tools will play an important role in future healthcare systems.

When thinking of the most common diseases in industrialised countries, such as cardiovascular diseases or diabetes, information exchange networks usually need to cover only a rather small geographic area. In case of less frequent or even rare conditions, however, health informatics and eHealth are becoming more and more connected across Europe. European Reference Networks for rare diseases are currently being implemented by the EU member states. Tools such as virtual tumour boards for oncological treatments or, generally spoken, virtual multidisciplinary meetings or conferences will play an important role within these networks. The Reference Networks, once fully functional, will not only ease access to specialised experts for patients with rare diseases, they will also build a European ICT infrastructure that might influence health care processes for the next decades. Therefore, it is essential that such infrastructures build on well-established standards such as Integrating the Healthcare Enterprise (IHE), even if standardization might initially require more time for implementation.

The 2017 issue of the annual Health Informatics meets eHealth conference addresses this increasingly international focus of eHealth and acknowledges the importance of cross-border health ICT. “Digital Insight – information-driven Health & Care” has been chosen as the special topic for eHealth2017 since future ICT systems need to include methods of machine learning and predictive analytics in order to provide actionable information to health professionals and to support preventive health care concepts, as described above. The present book, available as an open access eBook, was compiled to give some digital insight into the current research in health informatics and eHealth, addressing future issues of health and care.

March 21st, 2017

Graz, Vienna

Dieter Hayn

Günter Schreier

Background: Critical incident reporting systems (CIRS) are used as a means to collect anonymously entered information of incidents that occurred for example in a hospital. Analyzing this information helps to identify among others problems in the workflow, in the infrastructure or in processes.

Objectives: The entire potential of these sources of experiential knowledge remains often unconsidered since retrieval of relevant reports and their analysis is difficult and time-consuming, and the reporting systems often do not provide support for these tasks. The objective of this work is to develop a method for retrieving reports from the CIRS related to a specific user query.

Methods: atural language processing (NLP) and information retrieval (IR) methods are exploited for realizing the retrieval. We compare standard retrieval methods that rely upon frequency of words with an approach that includes a semantic mapping of natural language to concepts of a medical ontology.

Results: By an evaluation, we demonstrate the feasibility of semantic document enrichment to improve recall in incident reporting retrieval. It is shown that a combination of standard keyword-based retrieval with semantic search results in highly satisfactory recall values.

Conclusion: In future work, the evaluation should be repeated on a larger data set and real-time user evaluation need to be performed to assess user satisfactory with the system and results.

Background: Tagging text data with codes representing biomedical concepts plays an important role in medical data management and analysis. A problem occurs if there are ambiguous words linked to several concepts.

Objectives and Methods: This study aims at investigating word sense disambiguation based on word embedding and recurrent convolutional neural networks. The study focuses on terms mapped to multiple concepts of the Unified Medical Language System (UMLS).

Results: We created 20 text processing pipelines trained on a subset of the MeSH Word Sense Disambiguation (MSH WSD) data set, each pipeline disambiguating the sense of one word. The pipelines were then tested on a disjoint subset of MSH WSD data. Most pipelines achieved good or even excellent results (70% of the pipelines achieved at least 90% accuracy, 40% achieved at least 98% accuracy). One poor-performing outlier was detected.

Conclusion: The proposed approach can serve as a basis for an up-scaled system combining pipelines for many ambiguous words. The methods used here recently proved very successful in other fields of text understanding and can be expected to scale-up with improved availability of training data.

Background: The diagnosis – and hence definitions – of healthcare-associated infections (HAIs) rely on microbiological laboratory test results in specific constellations.

Objectives: To construct a library that provides interoperable building blocks for the analysis of microbiological laboratory test results.

Methods: We used Java for preprocessing raw microbiological laboratory test results and Arden Syntax for knowledge-based querying of data based on microbiology information elements used in European surveillance criteria for HAIs. To test the library and quantify how often these information elements occur in the data, we performed a retrospective cohort study on adult patients admitted for at least 24 hours to an intensive care unit at the Vienna General Hospital in 2013.

Results: We identified eleven information elements for which information was electronically available. These elements were identified positively 1,239 times in 1,184 positive microbiology tests from 563 patients.

Discussion: The availability of a library for the analysis of microbiology laboratory test results in HAI terms facilitates electronic HAI surveillance.

Routine patient data in electronic patient records are only partly structured, and an even smaller segment is coded, mainly for administrative purposes. Large parts are only available as free text. Transforming this content into a structured and semantically explicit form is a prerequisite for querying and information extraction. The core of the system architecture presented in this paper is based on SAP HANA in-memory database technology using the SAP Connected Health platform for data integration as well as for clinical data warehousing. A natural language processing pipeline analyses unstructured content and maps it to a standardized vocabulary within a well-defined information model. The resulting semantically standardized patient profiles are used for a broad range of clinical and research application scenarios.

Delirium is an acute confusion condition, which is common in elderly and often misdiagnosed in hospitalized patients. Early identification and prevention of delirium could reduce morbidity and mortality rates in those affected and reduce hospitalization costs. We have developed and validated a multivariate prediction model that predicts delirium and gives an early warning to physicians. A large set of patient electronic medical records have been used in developing the models. Classical learning algorithms have been used to develop the models and compared the results. Excellent results were obtained with the feature set and parameter settings attaining accuracy of 84%.

Among different used healthcare terminology resources there is a need to link them to ease their interoperability. SNOMED CT is the most detailed clinical reference terminology for procedure used in the Electronic Health Record (EHR). WHO developed since 2006 an aggregated classification named ICHI required for statistics and resource allocation. It is based on an ontology framework defined in ISO 1828 named Categorial Structure (CAST) for surgical procedures. We present ICHI coding structure, ISO 1828 CAST standard for surgical procedures and SNOMED CT procedures hierarchy concept model and their relations. We demonstrate the obstacles to align ICHI coding structure with ISO 1828 CAST which can be linked with SNOMED CT procedures hierarchy concept model. We recommend to update ICHI coding structure to decrease the gap with ISO 1828 CAST and SNOMED CT concept model for procedures hierarchy to allow users to share the different terminology resources.

Background: German university hospitals have started to establish molecular tumor boards in order to enable physicians to make molecular-guided decisions.

Objective: Our aim was to describe the organizational structure and procedures which are currently supporting the molecular tumor boards of five German university hospitals.

Methods: We conducted semi-structured interviews with experts of five university hospitals between December 2016 and February 2017.

Results: We observed heterogeneity in both the organization of genetic testing and the management of the molecular tumor boards among the five hospitals. They used free-text documents in most of their support procedures rather than machine-readable documents.

Conclusion: There are three potentialities to support the process from genetic testing to reporting within the molecular tumor boards: (i) standardized pipeline to integrate automated variant calling and annotation; (ii) tools supporting the experts in creating their reports and presentations and (iii) implementing pharmacogenomic CDSS into clinical routine.

Tele-collaboration between medical professionals is well established for specialties like radiology, cardiology or pathology. Typically these applications do not allow for ad hoc participation of non-registered users like patients and / or medical professionals. This paper describes an approach for extending these tele-applications to non-registered users and providing ad hoc participation with means for providing and accessing data. A light weight, web-based approach is used to assure acceptance and ease of use while maintaining and complying to the required regulations regarding authentication, authorization and protection of personal data. The approach has been implemented and is in use with a nationwide tele-collaboration network in Germany.

Background: Multidisciplinary team meetings (MDTMs) are already in use for certain areas in healthcare (e.g. treatment of cancer). Due to the lack of common standards and accessibility for the applied IT systems, their potential is not yet completely exploited.

Objectives: Common requirements for MDTMs shall be identified and aggregated into a process definition to be automated by an application architecture utilizing modern standards in electronic healthcare, e.g. HL7 FHIR.

Methods: To identify requirements, an extensive literature review as well as semi-structured expert interviews were conducted.

Results: Results showed, that interoperability and flexibility in terms of the process are key requirements to be addressed. An architecture blueprint as well as an aggregated process definition were derived from the insights gained. To evaluate the feasibility of identified requirements, methods of explorative prototyping in software engineering were used.

Conclusion: MDTMs will become an important part of modern and future healthcare but the need for standardization in terms of interoperability is imminent.

Background: For translational research, software platforms such as tranSMART with an integrated view of both clinical and omics data have gained more and more attention in the last years.

Objectives: We wanted to examine the success and failures of tranSMART in the fields of translational research and education by looking at the examples of six use cases at our hospital. We wanted to point out suitable scenarios and user groups as well as still existing challenges and limitations.

Methods: We sum up the experience we made with our use cases with a focus on lessons learned.

Results: tranSMART was successfully established by a bottom-up approach at our university hospital and has been running for more than two years now. It has been used in four translational research projects as well as in education in the context of lectures and bachelor/master theses.

Conclusion: tranSMART can be a very useful tool for translational research and education. But it should be used with both care and statistical knowledge to avoid wrong conclusions. Some technical constraints, especially for data modeling, still limit many applications. Version control and data provenance are remaining challenges.

Background: RESTful web services nowadays are state-of-the-art in business transactions over the internet. They are however not very much used in medical informatics and in clinical research, especially not in Europe.

Objectives: To make an inventory of RESTful web services that can be used in medical informatics and clinical research, including those that can help in patient empowerment in the DACH region and in Europe, and to develop some new RESTful web services for use in clinical research and regulatory review.

Methods: A literature search on available RESTful web services has been performed and new RESTful web services have been developed on an application server using the Java language.

Results: Most of the web services found originate from institutes and organizations in the USA, whereas no similar web services could be found that are made available by European organizations. New RESTful web services have been developed for LOINC codes lookup, for UCUM conversions and for use with CDISC Standards.

Discussion: A comparison is made between “top down” and “bottom up” web services, the latter meant to answer concrete questions immediately. The lack of RESTful web services made available by European organizations in healthcare and medical informatics is striking.

Conclusion: RESTful web services may in short future play a major role in medical informatics, and when localized for the German language and other European languages, can help to considerably facilitate patient empowerment. This however requires an EU equivalent of the US National Library of Medicine.

Background: Electronic collection and high quality analysis of medical data is expected to have a big potential to improve patient care and medical research. However, the integration of data from different stake holders is posing a crucial problem. The exchange and reuse of medical data models as well as annotations with unique semantic identifiers were proposed as a solution.

Objectives: Convert metadata from the Study of Health in Pomerania to the standardized CDISC ODM format.

Methods: The structure of the two data formats is analyzed and a mapping is suggested and implemented.

Results: The metadata from the Study of Health in Pomerania was successfully converted to ODM. All relevant information was included in the resulting forms. Three sample forms were evaluated in-depth, which demonstrates the feasibility of this conversion.

Conclusion: Hundreds of data entry forms with more than 15.000 items can be converted into a standardized format with some limitations, e.g. regarding logical constraints. This enables the integration of the Study of Health in Pomerania metadata into various systems, facilitating the implementation and reuse in different study sites.

The new virtual reality based medical applications is providing a better understanding of healthcare related subjects for both medical students and physicians. The work presented in this paper underlines gamification as a concept and uses VR as a new modality to study the human skeleton. The team proposes a mobile Android platform application based on Unity 5.4 editor and Google VR SDK. The results confirmed that the approach provides a more intuitive user experience during the learning process, concluding that the gamification of classical medical software provides an increased interactivity level for medical students during the study of the human skeleton.

Health information management (HIM) professionals have a combination of skills and, at the same time, the demand for their skills in the health system is increasing rapidly. This study aimed to assess the educational needs of the HIM staff in Iran. This descriptive analytical study was conducted in eight teaching hospitals. It was found that the maximum educational needs concerned the knowledge of medical terminology, occupational safety, legal aspects, the newest rules and regulations, and ministry guidelines, while the least of the felt needs related to insurance and other aspects of registry, data ownership, and data quality. The need to learn about coding and classifications had a significant relationship with work experience (P = 0.045) and those with a work experience of 6 to 10 years had fewer needs. Educational needs were also significantly associated with the number of years since graduation (P = 0.005), as those with 5–10 years' experience after post-graduation had lesser needs than others. Those who plan educational programs for health information professionals must have a comprehensive view of the needs of the health system. Participation of specialists of different fields must be considered in educational planning of such interdisciplinary fields.

Background: This paper presents a case study of the nationwide implementation of an Electronic Health Record system based on industry standards.

Objectives: To critically review the finalized nationwide EHR implementation in Albania, pointing out the achievements and the shortcomings, gained knowledge and sharing lessons learned.

Methods: This paper reports on our analysis of the project's documentation, first-hand experience working with healthcare professionals during and after project implementation. The initial system uptake analysis has been made using EHR usage statistics.

Results: Despite very poor initial ICT infrastructure, minimal or non-existing medical nomenclature and a rather challenging project schedule, the nationwide implementation of the EHR system was successfully completed in less than two years.

Conclusion: Although the system is already in use by 79 healthcare providers and is covering secondary and tertiary care, the majority of clinical data still remains in paper format; the access for clinicians to the system is limited due to insufficient ICT infrastructure.

While pharmacogenomic testing combined with clinical decision support has the potential to increase the safety and efficacy of medical treatments, the intake of multiple prescription drugs can – if not sufficiently addressed by decision support solutions – impair the effectiveness of such interventions by modulating the capacity of precisely those enzymes whose function pharmacogenomic tests try to predict. We quantified the potential extent of such drug-mediated mismatches between genotype-derived phenotypes and real phenotypes, commonly called “phenoconversion”, by screening claims data from 1,587,829 Austrian health insurance holders of the years 2006 and 2007 for concomitant prescriptions of drugs that can be dosed based on pharmacogenomics, and drugs that modulate enzyme activity. In total, 232,398 such prescription overlaps were detected, of which more than half (54.6%) could be attributed to co-prescriptions of moderate or strong modulators. Our results indicate that prescription drug-mediated phenoconversion is not uncommon, and should therefore be adequately reflected in decision support solutions by integrating algorithms to detect potential gene-drug-drug interactions.

Background: In Germany, adverse drug reactions and events cause hospitalizations, which lead to numerous thousands of deaths and several million Euros in additional health costs annually.

Objectives: Approximately one in two deaths could be avoided by an appropriate system for risk analysis of drugs.

Methods: The integration and storage of several data sources from life sciences are an ongoing need to address various questions with respect to drug therapy. A software architecture for data integration was implemented in order to build up a new data warehouse named KALIS-DWH, which includes pharmacological, biomolecular and patient-related data.

Results: Based on this comprehensive KALIS-DWH, an eHealth system named KALIS for biomedical risk analysis of drugs was implemented. The task-specific modules of KALIS offer efficient algorithms for analyzing medication and supporting decision-making in drug therapy.

Conclusion: KALIS is meant to be a web-based information system for health professionals and researchers. KALIS provides comprehensive knowledge and modules for risk analysis of drugs, which can contribute to minimizing prescribing errors.

Interoperability is a key requirement for any IT-System to be future proof and cost efficient, due to the increasing interaction of IT-Systems in Healthcare. This feasibility study is part of a larger project focusing on the conceptualization and evaluation of interoperable and modular IT-Framework components for exchanging big data information sets. Hence, this project investigates the applicability of a standard based IT-Architecture for the integration of Personal Health Devices data and open data sources. As a proof of concept use case, pollen forecast data from the Medical University of Vienna were combined with Personal Health Device data and a data correlation was investigated. The standards were identified as well as selected in expert's reviewed and the Architecture was designed based on a literature research. Subsequently the prototype was implemented and successfully tested in interoperability tests. The study shows that the architecture meets the requirements. It can be flexibly extended according to further requirements due to its generic setup. However, further extensions of the Interoperability-Connector and a full test setup needs to be realized in future.

Background: Tele-rehabilitation at home is one of the promising approaches in increasing rehabilitative success and simultaneously decreasing the financial burden on the healthcare system.

Objectives: Novel and mostly mobile devices are already in use, but shall be used in the future to a higher extent for allowing at home rehabilitation processes at a high quality level. The combination of exercises, assessments and available equipment is the basic objective of the presented database.

Methods: The database has been structured in order to allow easy-to-use and fast access for the three main user groups. Therapists – looking for exercise and equipment combinations – patients – rechecking their tasks for home exercises – and manufacturers – entering their equipment for specific use cases.

Results: The database has been evaluated by a proof of concept study and shows a high degree of applicability for the field of rehabilitative medicine. Currently it contains 110 exercises/assessments and 111 equipment/systems.

Conclusion: Foundations of presented database are already established in the rehabilitative field of application, but can and will be enhanced in its functionality to be usable for a higher variety of medical fields and specifications.

A variety of acute and chronic diseases require rehabilitation at home after treatment. Outpatient rehabilitation is crucial for the quality of the medical outcome but is mainly performed without medical supervision. Non-Compliance can lead to severe health risks and readmission to the hospital. While the patient is closely monitored in the hospital, methods and technologies to identify risks at home have to be developed. We analyze state-of-the-art monitoring systems and technologies and show possibilities to transfer these technologies into rehabilitation monitoring. For this purpose, we analyze sensor technology from the field of Quantified Self and Smart Homes. The available sensor data from this consumer grade technology is summarized to give an overview of the possibilities for medical data analytics. Subsequently, we show a conceptual roadmap to transfer data analytics methods to sensor based rehabilitation risk management.

Increasing treatment costs of HF patients affect the initiation of appropriate treatment method. Divergent approaches to measure the costs of treatment and the lack of common cost indicators impede the comparison of therapy settings.

Objectives: In the context of the present meta-analysis, key cost indicators from the perspective of healthcare providers are to be identified, described, analyzed and quantified. This review helps narrowing down the cost indicators, which have the most significant economic impact on the total treatment costs of HF patients. Telemedical services are to be compared to standard therapy methods.

Methods: The identification process was based on several steps. For the quantitative synthesis, we used the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. An additional set of criteria was defined for the following qualitative analysis.

Results: 5 key cost indicators were identified with significant economic impact on the treatment costs of HF patients.

Conclusion: 95% of the reported treatment costs could be captured based on the identified cost indicators.

The POSEIDON project aimed to increase the independence and autonomy of people with Down syndrome with the help of technical assistants. It followed a user-centered approach by involving people with Down syndrome and their parents, carers etc. A requirement analysis was the first step of the project. It became clear that people with Down syndrome especially need support in the areas of time management, mobility and money handling. Different applications were developed which were tested and evaluated in two field tests in three countries. Results indicate that POSEIDON can help to overcome daily challenges and that it can increase the autonomy and independence of people with Down syndrome.

Standard toilets often do not meet the needs of a significant number of older persons and persons with disabilities. The EU funded iToilet project aims at design and development of a new type of ICT enhanced modular toilet system which shall be able to support autonomy, dignity and safety of older persons living at home. Methodologically the project started with gathering user requirements by means of questionnaires, interviews and focus group discussion involving a total of 74 persons, thereof 41 subjects with movement disorders (primary users), 21 caregivers (secondary users) and 12 healthcare managers (tertiary users). Most important wishes were bilateral removable handrails, height and tilt adjustment, emergency detection, simplicity. In parallel to the ongoing technical development participatory design activities have been carried out at user test sites in order to continuously involve users into the design process and to allow quick feedback with regards to early prototype parts. The project currently is working on the finalization of the first prototype ready to enter the lab trial stage in spring 2017. The experiences will be used for redesigning a prototype 2 which is planned to be tested in real life settings early 2018.

Since 2012 six AAL pilot regions were launched in Austria. The main goal of these pilot regions is to evaluate the impact of AAL technologies in daily use considering the entire value chain. Additionally, go-to market strategies for assistive technologies based on an involvement of all relevant stakeholders are developed. Within this paper an overview of the specific objectives, approaches and the status of all Austrian AAL pilot regions is given. Taking into account the different experiences of the different pilot regions, specific challenges in establishing, implementing and sustaining pilot region projects are discussed and lessons-learned are presented. Results show that a careful planning of all project phases taking into account available resources is crucial for the successful implementation of an AAL pilot region. In particular, this applies to all activities related to the active involvement of end-users.