Ebook: Health Informatics Meets eHealth
Biomedical engineering and health informatics are closely related to each other, and it is often difficult to tell where one ends and the other begins, but ICT systems in healthcare and biomedical systems and devices are already becoming increasingly interconnected, and share the common entity of data. This is something which is set to become even more prevalent in future, and will complete the chain and flow of information from the sensor, via processing, to the actuator, which may be anyone or anything from a human healthcare professional to a robot. Methods for automating the processing of information, such as signal processing, machine learning, predictive analytics and decision support, are increasingly important for providing actionable information and supporting personalized and preventive healthcare protocols in both biomedical and digital healthcare systems and applications.
This book of proceedings presents 50 papers from the 12th eHealth conference, eHealth2018, held in Vienna, Austria, in May 2018. The theme of this year’s conference is Biomedical Meets eHealth – From Sensors to Decisions, and the papers included here cover a wide range of topics from the field of eHealth.
The book will be of interest to all those working to design and implement healthcare today.
Biomedical Meets eHealth – From Sensors to Decisions
Already todaybut even more in the futureICT systems in healthcarebiomedical systems and devices will increasingly be intertwined and share a common entity, which is data. This completes the chain and flow of information from the sensor via the processing to the actuator, which can be anything from a human healthcare professional to a robot. Along this pathway, methods for automating the information processing, like signal processing, machine learning, predictive analytics and decision support, play an increasing role to provide actionable information and to support personalized and preventive health care concepts, in both biomedical and digital healthcare systems and applications.
Both scientific disciplines, i.e. biomedical engineering and health informatics, are also closely related to each other and it is often difficult to delineate where the one ends and the other begins.
This starts with the practical settings, for example, in hospitals. Traditionally, there are two different organizational entities working together, i.e. the healthcare engineering and the ICT departments. The first primarily take care of the “hardware”, i.e. appliances, devices and systems, for example imaging equipment like Ultrasound machines or CT scanners. The latter take care of conveying the data generated by these items to the bedside and the healthcare specialists via Health Information Systems (HIS), Radiology Information Systems (RIS) or Picture Archiving and Communication Systems (PACS). However, these systems are interrelated and – as an example – it becomes more and more difficult to locate errors when combined systems come down. In the end, security is also a common concern and standards like the IEC 80001: Application of risk management for IT-networks incorporating medical devices address these overarching needs.
A look to the international level reveals, for example, that the IEEE Engineering in Medicine and Biology Society (EMBS),
Finally, since its beginning in 2007, the scientific backbone of our annual conference has been the working group for “Medical Informatics and eHealth” of the Austrian Society of Biomedical Engineering (OEGBMT)
Graz, March 2018
Logistic regression based propensity score matching is a widely used method in case-control studies to select the individuals of the control group. This method creates a suitable control group if all factors affecting the output variable are known. However, if relevant latent variables exist as well, which are not taken into account during the calculations, the quality of the control group is uncertain. In this paper, we present a statistics-based research in which we try to determine the relationship between the accuracy of the logistic regression model and the uncertainty of the dependent variable of the control group defined by propensity score matching. Our analyses show that there is a linear correlation between the fit of the logistic regression model and the uncertainty of the output variable. In certain cases, a latent binary explanatory variable can result in a relative error of up to 70% in the prediction of the outcome variable. The observed phenomenon calls the attention of analysts to an important point, which must be taken into account when deducting conclusions.
Lately, several studies started to investigate the existence of links between cannabis use and psychotic disorders. This work proposes a refined Machine Learning framework for understanding the links between cannabis use and 1st episode psychosis. The novel framework concerns extracting predictive patterns from clinical data using optimised and post-processed models based on Gaussian Processes, Support Vector Machines, and Neural Networks algorithms. The cannabis use attributes' predictive power is investigated, and we demonstrate statistically and with ROC analysis that their presence in the dataset enhances the prediction performance of the models with respect to models built on data without these specific attributes.
Background: Evidence-based clinical guidelines have a major positive effect on the physician's decision-making process. Computer-executable clinical guidelines allow for automated guideline marshalling during a clinical diagnostic process, thus improving the decision-making process.
Objectives: Implementation of a digital clinical guideline for the prevention of mother-to-child transmission of hepatitis B as a computerized workflow, thereby separating business logic from medical knowledge and decision-making.
Methods: We used the Business Process Model and Notation language system Activiti for business logic and workflow modeling. Medical decision-making was performed by an Arden-Syntax-based medical rule engine, which is part of the ARDENSUITE software.
Results: We succeeded in creating an electronic clinical workflow for the prevention of mother-to-child transmission of hepatitis B, where institution-specific medical decision-making processes could be adapted without modifying the workflow business logic.
Conclusion: Separation of business logic and medical decision-making results in more easily reusable electronic clinical workflows.
Background: Medical ward rounds are critical focal points of inpatient care that call for uniquely flexible solutions to provide clinical information at the bedside. While this fact is undoubted, adoption rates of mobile IT solutions remain rather low.
Objectives: Our goal was to investigate if and how mobile IT solutions influence successful information provision at the bedside, i.e. clinical information logistics, as well as to shed light at socio-organizational factors that facilitate adoption rates from a user-centered perspective.
Methods: Survey data were collected from 373 medical and nursing directors of German, Austrian and Swiss hospitals and analyzed using variance-based Structural Equation Modelling (SEM).
Results: The adoption of mobile IT solutions explains large portions of clinical information logistics and is in itself associated with an organizational culture of innovation and end user participation.
Conclusion: Results should encourage decision makers to understand mobility as a core constituent of information logistics and thus to promote close end-user participation as well as to work towards building a culture of innovation.
Mobile technologies have a positive impact on patient care and cause to improved decision making, reduced medical errors and improved communication in care team. The purpose of this study was to investigate the use of mobile technologies by medical and nursing students and their tendency in future. This study was conducted among 372 medical and nursing students of Tehran University of Medical Science. Respectively, 60.8% and 62.4% of medical and nursing students use smartphone. The most commonly used apps among medical students were medical dictionary, drug apps, medical calculators and anatomical atlases and among nursing students were medical dictionary, anatomical atlases and nursing care guides. Also, the use of decision support systems, remote monitoring, patient imagery and remote diagnosis, patient records documentation, diagnostic guidelines and laboratory tests will be increased in the future.
Background: IT is getting an increasing importance in hospitals. In this context, major IT decisions are often made by CEOs who are not necessarily IT experts.
Objectives: Therefore, this study aimed at a) exploring different types of IT decision makers at CEO level, b) identifying hypotheses if trust exists between these different types of CEOs and their CIOs and c) building hypotheses on potential consequences regarding risk taking and innovation.
Methods: To this end, 14 qualitative interviews with German hospital CEOs were conducted to explore the research questions.
Results: The study revealed three major types: IT savvy CEOs, IT enthusiastic CEOs and IT indifferent CEOs. Depending on these types, their relationship with the CIO varied in terms of trust and common language. In case of IT indifferent CEOs, a potential vicious circle of lack of IT knowledge, missing trust, low willingness to take risks and low innovation power could be identified.
Conclusion: In order to break of this circle, CEOs seem to need more IT knowledge and/or greater trust in their CIO.
Background: The calculation of daily fluid balances is essential in perioperative and postoperative fluid management in order to prevent severe hypervolemia or hypovolemia in critically ill patients. In this context, modern health information technology has the potential to reduce the workload for health care professionals by not only automating data collection but also providing appropriate decision support.
Objectives: Within this work, current problems and barriers regarding fluid balancing in cardiac intensive care patients are outlined and improvement activities are specified.
Methods: Literature research and qualitative interviews with health care professionals were conducted to assess the state-of-the-art technological setting within an intensive care unit.
Results: An example case shows that interconnecting not only devices but also wards can facilitate daily clinical tasks.
Conclusion: Smart devices and decision support systems can improve fluid management. Several technologies, which today are sometimes still considered to be futuristic, are in fact not that far away or already available. However, they need proper implementation with respect to intensivists', nurses' and patients' needs.
Background: The development processes of data exchange standards for use in healthcare are very different from those used in clinical research. Healthcare data standards are traditionally developed by the Health Level 7 (HL7) organization, whereas those for use in clinical research are mostly developed by the Clinical Data Interchange Standards Consortium. No alignment of these standards has so far taken place.
Objectives: Due to the increasing use of electronic health records as primary source in clinical research, it becomes necessary to align these standards, not only the semantic standards, but also the data exchange standards (formats) themselves.
Methods: Mutual feature gaps between ODM and FHIR are investigated.
Results: A transition path how the HL7-FHIR standard and the CDISC-ODM transport standard can grow into a single standard for use both in healthcare and in clinical research is presented.
Background: The research project REPO (Radiology Ehealth PlatfOrm) was started in 2017 with the goal to “enable cross-enterprise collaboration in radiology using the Austrian eHealth infrastructure”.
Objectives: The objective of this paper was to provide an overview of the radiology IT environment – actors, use cases and technology.
Methods: We conducted semi-structured expert interviews with radiologists and hospital operators and we statistically analyzed the client database of our research project partner.
Results: Interviews led to a list of use cases where cross-enterprise collaboration in radiology takes place and the data analysis provided insights on the systems, networks and standards in place.
Conclusion: The Austrian IT infrastructure in radiology is a heterogeneous naturally grown environment. Future developments should be based on internationally accorded standards and on integration profiles provided by Integrating the Healthcare Enterprise (IHE).
Background: Discharge summaries are a standard communication tool delivering important clinical information from inpatient to ambulatory care. To ensure a high quality, correctness and completeness, the generation process is time consuming. It requires also contributions of multiple persons. This is problematic since the primary care provider needs the information from the discharge summary for continuing the intended treatment. To address this challenge, we developed a concept for exchanging a modular electronic discharge summary.
Methods: Through a literature review and interviews with multiple stakeholders, we analysed existing processes and derived requirements for an improved communication of the discharge summary.
Results: In this paper, we suggest a concept of a modular electronic discharge summary that is exchanged through the electronic patient dossier in CDA CH level 2 documents. Until 2020, all Swiss hospitals are obliged to connect to the electronic patient dossier. Our concept allows to access already completed modules of the discharge summary from the primary care side, before the entire report is entirely finalised. The data is automatically merged with the local patient record on the physician side and prepared for data integration into the practice information system.
Conclusion: Our concept offers the opportunity not only to improve the information exchange between hospital and primary care, but it also provides a potential use case and demonstrates a benefit of the electronic patient dossier for primary care providers who are so far not obliged to connect to the patient dossier in Switzerland.
Background: An ever growing for application of electronic health records (EHRs) has improved healthcare providers' communications, access to data for secondary use and promoted the quality of services. Patient's privacy has been changed to a great issue today since there are large loads of critical information in EHRs. Therefore, many privacy preservation techniques have been proposed and anonymization is a common one.
Objectives: This study aimed to investigate the effectiveness of anonymization in preserving patients' privacy.
Methods: The articles published in the 2005–2016 were included. Pubmed, Cochrane, IEEE and ScienceDirect were searched with a variety of related keywords. Finally, 18 articles were included.
Results: In the present study, the relevant anonymization issues were investigated in four categories: secondary use of anonymized data, re-identification risk, anonymization effect on information extraction and inadequacy of current methods for different document types.
Conclusion: The results revealed that though anonymization cannot reduce the risk of re-identification to zero, if implemented correctly, can manage to help preserve patient's privacy.
Data Warehouses (DW) are useful tools to support clinical studies as they can provide exports of routine care data for scientific reuse. Exported DW data is usually post-processed and integrated into study databases by study staff that is reasonably trained in specific tools like SPSS and Excel but which are no programmers or computer scientists. DW systems should therefore be configurable to satisfy export format desiderata as much as possible so that exports contain no unnecessary post-processing obstacles. In the presented work the authors analyze various existing DW systems in respect to a list of potential export formats.
Background: To develop simulation models for healthcare related questions clinical data can be reused.
Objectives: Develop a clinical data warehouse to harmonize different data sources in a standardized manner and get a reproducible interface for clinical data reuse.
Methods: The Kimball life cycle for the development of data warehouse was used. The development is split into the technical, the data and the business intelligence pathway.
Results: Sample data was persisted in the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM). The i2b2 clinical data warehouse tools were used to query the OMOP CDM by applying the new i2b2 multi-fact table feature.
Conclusion: A clinical data warehouse was set up and sample data, data dimensions and ontologies for Austrian health claims data were created. The ability of the standardized data access layer to create and apply simulation models will be evaluated next.
Patients with multiple disorders usually have long diagnosis lists, constitute by ICD-10 codes together with individual free-text descriptions. These text snippets are produced by overwriting standardized ICD-Code topics by the physicians at the point of care. They provide highly compact expert descriptions within a 50-character long text field frequently not assigned to a specific ICD-10 code. The high redundancy of these lists would benefit from content-based categorization within different hospital-based application scenarios. This work demonstrates how to accurately group diagnosis lists via a combination of natural language processing and hierarchical clustering with an overall F-measure value of 0.87. In addition, it compresses the initial diagnosis list up to 89%. The manuscript discusses pitfall and challenges as well as the potential of a large-scale approach for tackling this problem.
Model-based decision support systems promise to be a valuable addition to oncological treatments and the implementation of personalized therapies. For the integration and sharing of decision models, the involved systems must be able to communicate with each other. In this paper, we propose a modularized architecture of dedicated systems for the integration of probabilistic decision models into existing hospital environments. These systems interconnect via web services and provide model sharing and processing capabilities for clinical information systems. Along the lines of IHE integration profiles from other disciplines and the meaningful reuse of routinely recorded patient data, our approach aims for the seamless integration of decision models into hospital infrastructure and the physicians' daily work.
Background: A challenge of using electronic health records for secondary analyses is data quality. Body mass index (BMI) is an important predictor for various diseases but often not documented properly.
Objectives: The aim of our study is to perform data cleansing on BMI values and to find the best method for an imputation of missing values in order to increase data quality. Further, we want to assess the effect of changes in data quality on the performance of a prediction model based on machine learning.
Methods: After data cleansing on BMI data, we compared machine learning methods and statistical methods in their accuracy of imputed values using the root mean square error. In a second step, we used three variations of BMI data as a training set for a model predicting the occurrence of delirium.
Results: Neural network and linear regression models performed best for imputation. There were no changes in model performance for different BMI input data.
Conclusion: Although data quality issues may lead to biases, it does not always affect performance of secondary analyses.
Delirium is an acute neuropsychiatric syndrome which is common in elderly patients during their hospitalisation and is associated with an increased mortality and morbidity. Since delirium is a) often underdiagnosed and b) preventable if early signs are detected,igh expectations are set in delirium risk assessment during hospital admission. In our latest studies, we showed that delirium prediction using machine learning algorithms is possible based on the patients' health history. The aim of this study is to compare the influence of nursing assessment data on prediction models with clinical and demographic data. We approached the problem by a) comparing the performance of predictive models including nursing data with models based on clinical and demographic data only and b) analysing the feature importance of all available features. From our results we concluded that nursing assessment data can improve the performance of delirium prediction models better than demographic, laboratory, diagnosis, procedures, and previous transfers' data alone.
Background: A fast and accurate data transmission from glucose meter to clinical decision support systems (CDSSs) is crucial for the management of type 2 diabetes mellitus since almost all therapeutic interventions are derived from glucose measurements.
Objectives: Aim was to develop a prototype of an automated glucose measurement transmission protocol based on the Continua Design Guidelines and to embed the protocol into a CDSS used by healthcare professionals.
Methods: A literature and market research was performed to analyze the state-of-the-art and thereupon develop, integrate and validate an automated glucose measurement transmission protocol in an iterative process.
Results: Findings from literature and market research guided towards the development of a standardized glucose measurement transmission protocol using a middleware. The interface description to communicate with the glucose meter was illustrated and embedded into a CDSS.
Conclusion: A prototype of an interoperable transmission of glucose measurements was developed and implemented in a CDSS presenting a promising way to reduce medication errors and improve user satisfaction.
Background: Due to the widespread use of mobile technology and the low cost of this technology, implementing a mobile-based self-management system can lead to adherence to the medication regimens and promotion of the health of people living with HIV (PLWH). We aimed to identify requirements of a mobile-based self-management system, and validate them from the perspective of infectious diseases specialists.
Method: This is a mixed-methods study that carried out in two main phases. In the first phase, we identified requirements of a mobile-based self-management system for PLWH. In the second phase, identified requirements were validated using a researcher made questionnaire. The statistical population was infectious diseases specialists affiliated to Tehran University of Medical Sciences. The collected data were analyzed using SPSS statistical software (version 19), and descriptive statistics.
Results: By full-text review of selected studies, we determined requirements of a mobile-based self-management system in four categories: demographic, clinical, strategically and technical capabilities. According to the findings, 6 data elements for demographic category, 11 data elements for clinical category, 10 items for self-management strategies, and 11 features for technical capabilities were selected.
Conclusion: Using the identified preferences, it is possible to design and implement a mobile-based self-management system for HIV-positive people. Developing a mobile-based self-management system is expected to progress the skills of self-management PLWH, improve of medication regimen adherence, and facilitate communication with healthcare providers.
The purpose of this study was to review different telemedicine services in diagnosis, treatment and management of various children diseases and providing an overview of systematic reviews conducted in this regard. We searched English articles published in peer-reviewed journals between 2000 to 2016. We found that tele-pediatric services have been reported in various areas such as cardiology, burn, diabetes, obesity, emergency medicine, speech and hearing loss, Ear, Nose and Throat, psychology and psychiatry, radiology, oncology, home healthcare, asthma, genetics and dentistry. These studies mainly reported positive results. However, systematic reviews in tele-pediatric showed that these studies have not proven the clinical effectiveness or suggested further studies to assess the clinical outcomes of services provided through telemedicine technologies.
Existing full-body tracking systems, which use Inertial Measurement Units (IMUs) as sensing unit, require expert knowledge for setup and data collection. Thus, the daily application for human body tracking is difficult. In particular, in the field of active and assisted living (AAL), tracking human movements would enable novel insights not only into the quantity but also into the quality of human movement, for example by monitoring functional training. While the current market offers a wide range of products with vastly different properties, literature lacks guidelines for choosing IMUs for body tracking applications. Therefore, this paper introduces developments towards an IMU evaluation framework for human body tracking which compares IMUs against five requirement areas that consider device features and data quality. The data quality is assessed by conducting a static and a dynamic error analysis. In a first application to four IMUs of different component consumption, the IMU evaluation framework convinced as promising tool for IMU selection.
Background: Manual skills teaching, such as physiotherapy education, requires immediate teacher feedback for the students during the learning process, which to date can only be performed by expert trainers.
Objectives: A machine-learning system trained only on correct performances to classify and score performed movements, to identify sources of errors in the movement and give feedback to the learner.
Methods: We acquire IMU and sEMG sensor data from a commercial-grade wearable device and construct an HMM-based model for gesture classification, scoring and feedback giving. We evaluate the model on publicly available and self-generated data of an exemplary movement pattern executions.
Results: The model achieves an overall accuracy of 90.71% on the public dataset and 98.9% on our dataset. An AUC of 0.99 for the ROC of the scoring method could be achieved to discriminate between correct and untrained incorrect executions.
Conclusion: The proposed system demonstrated its suitability for scoring and feedback in manual skills training.
Background: Due to the chronicity of HIV/AIDS and the increased number of people living with HIV (PLWH), these people need the innovative and practical approaches to take advantage of high-quality healthcare services. The objectives of this scoping review were to identify the mobile-based applications and functionalities for self-management of people living with HIV.
Methods: We conducted a comprehensive search of PubMed, Scopus, Science direct, Web of Science and Embase databases for literature published from 2010 to 2017. Screening, data abstraction, and methodological quality assessment were done in duplicate.
Results: Our search identified 10 common mobile-based applications and 8 functionalities of these applications for self-management of people living with HIV. According to the findings, “text-messaging” and “reminder” applications were more addressed in reviewed articles. Moreover, the results indicated that “medication adherence” was the common functionality of mobile-based applications for PLWH.
Conclusion: Inclusive evidence supports the use of text messaging as a mobile-based functionality to improve medication adherence and motivational messaging. Future mobile-based applications in the healthcare industry should address additional practices such as online chatting, social conversations, physical activity intervention, and supply chain management.
Background: Medical device regulations which aim to ensure safety standards do not only apply to hardware devices but also to standalone medical software, e.g. mobile apps.
Objectives: To explore the effects of these regulations on the development and distribution of medical standalone software.
Methods: We invited a convenience sample of 130 domain experts to participate in an online survey about the impact of current regulations on the development and distribution of medical standalone software.
Results: 21 respondents completed the questionnaire. Participants reported slight positive effects on usability, reliability, and data security of their products, whereas the ability to modify already deployed software and customization by end users were negatively impacted. The additional time and costs needed to go through the regulatory process were perceived as the greatest obstacles in developing and distributing medical software.
Conclusion: Further research is needed to compare positive effects on software quality with negative impacts on market access and innovation. Strategies for avoiding over-regulation while still ensuring safety standards need to be devised.