Ebook: Navigating Healthcare Through Challenging Times

For more than a year, the COVID-19 crisis has been affecting healthcare systems worldwide. Aside from its various dramatic effects on citizens and healthcare professionals, COVID-19 has also triggered and accelerated some important process changes. Digital health has played a role of the utmost importance in the last 12 months, supporting contact tracing, test strategies, statistical analyses, prognostic modeling, vaccination organization and documentation, etc. Teleconferences have become common as a way of communicating, not only on a business level, but also in our daily personal lives, even between grandparents and their grandchildren, and it seems likely that these changes will influence healthcare even in the longer-term.

These proceedings provide an insight into research on digital health as conducted during the COVID-19 crisis, including articles concerning the management of infectious diseases, telehealth services, standardization and interoperability in healthcare, nursing informatics, data analytics, predictive modeling, digital tools for rare disease research, and many others.

The dHealth conference series has been contributing to scientific exchange and networking since 2007, and the conference was organized in a virtual setting for a second time in 2021. Each year, this event attracts 300+ participants from academia, industry, government and healthcare organizations. With its motto, “Health Informatics meets Digital Health”, the event provides a platform for researchers, practitioners, decision makers and vendors to discuss innovative health informatics and dHealth solutions with the aim of improving the quality and efficiency of healthcare. It is the USP of this event that it is where research and application meet as equals to provide new insights from both the scientific and practical points of view.

Dieter Hayn, Günter Schreier and Martin Baumgartner

Graz, Vienna, April 2021

The benefits of eHealth interventions for people with dementia and their informal caregivers have been demonstrated in several studies. In times of contact restrictions, digital solutions have become increasingly important, especially for people with dementia and their mostly elderly caregiving relatives, which are at increased risk of severe illness from COVID-19. As in many other health areas, there is a lack of digital interventions in the dementia landscape that are successfully implemented (i.e., put into practice), especially digital interventions that are scientifically evaluated. Evaluated and proven effective digital interventions exist, but these often do not find their way from research into practice and stay on low-level implementation readiness. Within the project digiDEM Bayern, a digital platform with digital services and interventions for people affected by dementia (people with dementia, caregivers, volunteers and interested citizens) is established. As one digital intervention for informal caregivers, the ‘Angehörigenampel’ (caregivers’ traffic-light) was developed, which is able to assess the physical and psychological burden of caregivers. This can help to counteract the health effects of caregiving burden early on before it is too late. The development of the digital intervention as a WordPress-plugin was kept generic so that it can easily be adapted to other languages on further websites. The ‘intervention as a plugin’ approach demonstrates an easy and flexible way of deploying eHealth interventions to other service providers, especially from other countries. The implementation barriers for other service providers are low enough for them to be able to easily integrate the eHealth intervention on their website, enabling more caregivers to benefit from the disseminated eHealth intervention.

Practice efficiency is influenced by its operations management. We aim at studying implementation of operations management in Swiss medical practices and we develop a dashboard that allows controlling and managing resources. To study operations management and relevant indicators in ambulant care, we distributed questionnaires by e-mail and conducted 6 interviews. In collaboration with a group practice, we collected requirements regarding a dashboard for operations management, developed a mockup and finally a prototype. This prototype was deployed and implemented in daily routine. From the assessments we learned that practice information systems (PIS) are not sufficiently supporting production planning and control. Relevant indicators include processing time per patient or waiting time for quantifying efficiency and identify potential improvements in production. Within 5 weeks of implementation of our dashboard in a group practice, we learned that calculating indicators and support of operations management by means of a dashboard is well appreciated by practice employees. Indicators are considered extremely useful for operations management.

Background:

Physicians spend a lot of time in routine tasks, i.e. repetitive and time consuming tasks that are essential for the diagnostic and treatment process. One of these tasks is to collect information on the patient’s medical history.

Objectives:

We aim at developing a prototype for an intelligent interviewer that collects the medical history of a patient before the patient-doctor encounter. From this and our previous experiences in developing similar systems, we derive recommendations for developing intelligent interviewers for concrete medical domains and tasks.

Methods:

The intelligent interviewer was implemented as chatbot using IBM Watson assistant in close cooperation with a family doctor.

Results:

AnCha is a rule-based chatbot realized as decision tree with 75 nodes. It asks a maximum of 44 questions on the medical history, current complaints and collects additional information on the patient, social details, and prevention.

Conclusion:

When developing an intelligent digital interviewer it is essential to define its concrete purpose, specify information to be collected, design the user interface, consider data security and conduct a practice-oriented evaluation.

Background:

Social networks are a good source for monitoring public health during the outbreak of COVID-19, these networks play an important role in identifying useful information.

Objectives:

This study aims to draw a comparison of the public’s reaction in Twitter among the countries of West Asia (a.k.a Middle East) and North Africa in order to make an understanding of their response regarding the same global threat.

Methods:

766,630 tweets in four languages (Arabic, English French, and Farsi) tweeted in March 2020, were investigated.

Results:

The results indicate that the only common theme among all languages is “government responsibilities (political)” which indicates the importance of this subject for all nations.

Conclusion:

Although nations react similarly in some aspects, they respond differently in others and therefore, policy localization is a vital step in confronting problems such as COVID-19 pandemic.

Background:

European universities offer a variety of programs in Medical Informatics. The Europen Federation of Medical Informatics (EFMI) offers accreditation of these programs.

Objectives:

To describe the process of EFMI accreditation of a new Austrian master’s program and reflect on accreditation benefits.

Methods:

Reflection on feedback and experiences at UMIT TIROL

Results:

Accreditation needs quite some preparation but offers essential opportunities for self-reflection and feedback by international experts.

Conclusion:

Besides national accreditation, medical informatics programs can benefit from the accreditation through international organizations.

Autism spectrum disorder (ASD) diagnoses increased over the last decades, as reviews show comparing prevalence rates reported from different studies. Due to different effects of the disorder, personal support is required and provided by formal and/or informal caregivers in various activities of daily living. With the help of a customized smart home and interior design concept the aim is to enable people with ASD to live a more independent and self-reliant life. Following a participatory research approach, the end users are involved in the context of use and requirements definition, concept development, and later also in the implementation, and evaluation process. The solution shall assist end users in performing activities of daily living. The outcome of the work at hand is a set of modular functionalities (sensors, actuators, interior design solutions) to be integrated in a living environment specifically designed for people with ASD.

Reuse of EHR data can substantially improve the recruitment process of clinical trials. As shown earlier, Shared EHR systems are particularly attractive data sources. The goal of this work was to conceptually design and implement a user-friendly tool for semiautomatic trial recruitment using ELGA data. The tool applies a web-based client (Vue and Electron frameworks) – server (Django-Python and Java server, SQLite database) architecture. Trial eligibility criteria are expressed as XPaths. Access to ELGA documents is simulated using the eHealth Connector library and the IHE XDS Open eHealth Integration Platform framework. Usability was optimized in expert interviews with investigators of two active trials. First feedback based on synthesized ELGA test data indicates suitability for clinical end users. Further insights are expected from applying the tool to real ELGA data.

Background:

Telerehabilitation represents a new cutting-edge method in the treatment of patients suffering from motor and cognitive disorders caused by stroke. Even if there exist dedicated devices able to track patients’ movements to evaluate the performed rehabilitation exercises, they require specific settings necessary for a correct and simple use at the patient’s home. If we consider the recent pandemic situation and the lockdown condition, which made difficult the access to these products, post stroke patients may be not able to perform home rehabilitation.

Objectives:

the goal of this work is the design of a specific method to develop a tele-rehabilitation platform for post-stroke patients using consumer technologies without involving ad-hoc devices.

Method:

Open-source tools have been investigated for speeding up the development starting with the medical knowledge.

Results:

a group of four healthcare technologies engineering students with no specific skills about computer science has developed a platform in four months using the design method.

Conclusion:

the presented method allowed the development of a clinical knowledge-based web platform for post-stroke patients totally based on consumer technology.

Hydrogen breath tests are a well-established method to help diagnose functional intestinal disorders such as carbohydrate malabsorption or small intestinal bacterial overgrowth. In this work we apply unsupervised machine learning techniques to analyze hydrogen breath test datasets. We propose a method that uses 26 internal cluster validation measures to determine a suitable number of clusters. In an induced external validation step we use a predefined categorization proposed by a medical expert. The results indicate that the majority of the considered internal validation indexes was not able to produce a reasonable clustering. Considering a predefined categorization performed by a medical expert, a novel shape-based method obtained the highest external validation measure in terms of adjusted rand index. The predefined clusterings constitute the basis of a supervised machine learning step that is part of our ongoing research.

In healthcare studies, the analysis of claims data is gaining an increasingly important role. Observational studies should be reported in a manner that promotes internal and external validity assessment, with the exact and standardized description of items. Several international guidelines and checklists for reporting on secondary data are available. The aim of this work was to analyse the applicability of reporting guidelines especially for claims data. The STROSA-2 guidelines recommendations were evaluated by means of a report on a study on triptan medications in Austria. Six items were identified which could be expanded to support complete and transparent report on Austrian claims data. Therefore, we would suggest to add some details in the STROSA-2 guidelines concerning study design, legal foundations, data protection, data flow, descriptive results and risk of bias. The guidelines for reporting on Austrian claims data were successfully compiled with additional items. New guidelines should be further processed and tested with strong recommendations to focus on data limitations and legal aspects.

Background:

Considering the impacts of the COVID-19 pandemic on health service delivery, the US Office for Civil Rights (OCR) updated the policies on health data processing, and Health Insurance Portability and Accountability Act (HIPAA).

Objectives:

In this study, we investigated discourses on HIPAA in relation to COVID-19.

Methods:

Through a search of media sources in the Factiva database, relevant texts were identified. We applied a text mining approach to identify concepts and themes in these texts.

Results:

Our analysis revealed six central themes, namely, Health, HIPAA, Privacy, Security, Patients, and Need, as well as their associated concepts. Among these, Health was the most frequently discussed theme. It comprised concepts such as public, care, emergency, providers, telehealth, entity, use, discretion, OCR, Health and Human Services (HHS), enforcement, business, and services.

Conclusion:

Our discourse analysis of media outlets highlights the role of health data privacy law in the response to global public health emergencies and demonstrates how discourse analysis and computational methods can inform health data protection policymaking in the digital health era.

Background:

Physical activity helps improve the overall quality of life. The correct execution of physical activity is crucial both in sports as well as disease prevention and rehabilitation. Little to no automated commodity solutions for automated analysis and feedback exist.

Objectives:

Validation of the Apple ARKit framework as a solution for automatic body tracking in daily physical exercises using the smartphones’ built-in camera.

Methods:

We deliver insights into ARKit’s body tracking accuracy through a lab experiment against the VICON system as Gold Standard. We provide further insights through case studies using apps built on ARKit.

Results:

ARKit exposes significant limitations in tracking the full range of motion in joints but accurately tracks the movement itself. Case studies show that applying it to measure the quantity of execution of exercises is possible.

Conclusion:

ARKit is a light-weight commodity solution for quantitative assessment of physical activity. Its limitations and possibilities in qualitative assessment need to be investigated further.

Care pathways and supporting health information systems (HIS) have been permeate the discipline of Health Information Systems Research (HISR) over years. Traditional objectives of workflow assistance are increasingly extended by interdisciplinary goals from technology, medicine, management and public health research. A systematic literature review is dedicated to this integrating character. It examines the interdisciplinary mesh of objectives associated with care pathways and pathway-supporting HIS in the HISR literature. From 47 identified articles, 6 thematic themes were derived. Their consolidation supports in particular design and development processes as it describes the solution space of future pathway-supporting HIS addressing requirements stated by multiple stakeholders.

Experiences of war and persecution often lead to mental health problems, resulting in post-traumatic stress disorders. In this work, we design a digital platform that aims at helping refugees coming to Switzerland by providing exercises for their mental health and information about daily life in Switzerland. In collaboration with the Swiss Red Cross (SRC), we collected requirements and developed a concept for information provision through in this platform. The architecture of a progressive web application (PWA) was identified as to best fulfill the given requirements. Based on the collected requirements mockups were created. In user interviews, we received feedback regarding the future system. We learned that the platform should include an avatar, which guides the user through the entire platform and asks questions. All texts should be accessible by a read-aloud function and exercises should be provided as videos. In summary, we learned that it is essential to involve the future user group in the development process since it is characterized by cultural diversity that has to be considered in the development and design. Enriched by this input, the next step is to realize the application in terms of a prototype.

Background:

Mobile apps may encourage a lifestyle that avoids unhealthy behaviors, such as smoking or poor nutrition, which promotes cardiovascular diseases (CVD). Yet, little data is available on the utilization, perception, and long-term effects of such apps to prevent CVD.

Objectives:

To develop a mobile app concept to reduce the individual CVD risk and collect information addressing research questions on CVD prevention while preserving data privacy and security.

Methods:

To validate the concept, a prototype will be built, and usability studies will be performed.

Results:

We expect to determine whether it is possible to reach a broad user base and to collect scientific information while protecting user data sufficiently.

Conclusion:

To address CVD prevention, we propose a mobile coaching app. We expect high acceptance rates in validation studies.

Background:

There is a lack of secure official communication channels for peer review and peer feedback on medical findings.

Objectives:

We aimed to utilize the existing Austrian eHealth infrastructure to enable review and feedback processes.

Methods:

We extended the IHE XDW workflow document to enable the exchange of text messages (i.e., comments on documents or images) over an XDS infrastructure.

Results:

The workflow enables the exchange of comments on specific sections of CDA documents or radiological images and was verified in an XDS test environment.

Conclusion:

The presented solution is a proof of concept and the potential basis for the specification of a new IHE workflow definition.

Reducing passenger flow through highly frequented bottlenecks in public transportation networks is a well-known urban planning problem. This issue has become even more relevant since the outbreak of the SARS-CoV-2 pandemic and the necessity for minimum distances between passengers. We propose an approach that allows to dynamically navigate passengers around dangerously crowded stations to better distribute the passenger load across an entire urban public transport network. This is achieved through the introduction of new constraints into routing requests, that enable the avoidance of specific nodes in a network. These requests consider walks, bikes, metros, subways, trams and buses as possible modes of transportation. An implementation of the approach is provided in cooperation with the Munich Travel Corporation (MVG) for the city of Munich, to simulate the effects on a real city’s urban traffic flow. Among other factors, the impact on the travel time was simulated given that the two major exchange points in the network were to be avoided. With an increase from 26.5 to 26.8 minutes on the average travel time, the simulation suggests that the time penalty might be worth the safety benefits.

Background:

Mobile-based social media play an important role in the dissemination of information during public health emergencies.

Objectives:

This study aimed to analyze the contents and trends of public messages posted on Telegram during Coronavirus Disease 2019 (COVID-19) pandemic.

Methods:

A content analysis of the 1781 messages, posted in a public Telegram channel with more than one million subscribers performed over 9-weeks. The messages were categorized into seven categories.

Results:

In total, 39% (n=703) of all messages were related to COVID-19. With the official confirmation of the case of COVID-19 in Iran, the number of COVID-related massages started to rise. Overall, the most frequent messages were of joke and humor (n=292, 41.5%), followed by educational messages (n=140, 19.9%).

Conclusion:

Our study showed that the most popular messages during first weeks of COVID pandemic were satirical, indicating that people may not had taken the risks of this pandemic seriously. It is crucial for health organizations to develop strategies for dissemination of reliable health information through social media.

Background:

Intensified research and innovation and rapid uptake of new tools, interventions, and strategies are crucial to fight Tuberculosis, the world’s deadliest infectious disease. The sharing of health data remains a significant challenge. Data consumers must be able to verify the consistency and integrity of data. Solutions based on distributed ledger technologies may be adequate, where each member in a network holds a unique credential and stores an identical copy of the ledger and contributes to the collective process of validating and certifying digital transactions.

Objectives:

This work proposes a mechanism and presents a use case in Digital Health to allow the verification of integrity and immutability of TB electronic health records.

Methods:

IOTA was selected as a supporting tool due to its data immutability, traceability and tamper-proof characteristics.

Results:

A mechanism to verify the integrity of data through hash functions and the IOTA network is proposed. Then, a set of TB related information systems was integrated with the network.

Conclusion:

IOTA technology offers performance and flexibility to enable a reliable environment for electronic health records.

Background:

Patients with major adverse cardiovascular events (MACE) such as myocardial infarction or stroke suffer from frequent hospitalizations and have high mortality rates. By identifying patients at risk at an early stage, MACE can be prevented with the right interventions.

Objectives:

The aim of this study was to develop machine learning-based models for the 5-year risk prediction of MACE.

Methods:

The data used for modelling included electronic medical records of more than 128,000 patients including 29,262 patients with MACE. A feature selection based on filter and embedded methods resulted in 826 features for modelling. Different machine learning methods were used for modelling on the training data.

Results:

A random forest model achieved the best calibration and discriminative performance on a separate test data set with an AUROC of 0.88.

Conclusion:

The developed risk prediction models achieved an excellent performance in the test data. Future research is needed to determine the performance of these models and their clinical benefit in prospective settings.

Background:

Integration of heterogenous resources is key for Rare Disease research. Within the EJP RD, common Application Programming Interface specifications are proposed for discovery of resources and data records. This is not sufficient for automated processing between RD resources and meeting the FAIR principles.

Objective:

To design a solution to improve FAIR for machines for the EJP RD API specification.

Methods:

A FAIR Data Point is used to expose machine-actionable metadata of digital resources and it is configured to store its content to a semantic database to be FAIR at the source.

Results:

A solution was designed based on grlc server as middleware to implement the EJP RD API specification on top of the FDP.

Conclusion:

grlc reduces potential API implementation overhead faced by maintainers who use FAIR at the source.

Background:

Delirium is a patient safety issue that often occurs within the population of elderly people. As delirium may be characterized by fluctuating progress, the aim of this work is to find methods to visualize the occurrence of delirium over time in different patient stays in gerontopsychatric settings.

Methods:

We analyzed current data mining visualization techniques for clinical research using a delirium data set collected in a gerontopsychatric setting.

Results:

We identified heatmaps and dendrograms resulting from hierarchical clustering as a suitable visualization method.

Conclusion:

Heat maps with hierarchical clustering are a suitable data mining tool or visualization technique to study delirium cases in the time course of patient stays.

Clinical decision support systems (CDSS) have been shown in a variety of diseases to lead to improvements in care. The aim of this study is to design a CDSS to assist GPs to assess and manage breathlessness, a highly prevalent symptom in practice. A focus group is conducted to explore the needs of general practitioners (GPs), assess current workflow to identify points for intervention and develop early prototypes for testing. Five GPs took part in the focus group elucidating 248 relevant data points which were then qualitatively analyzed using the Technology Acceptance Model as the theoretical framework. In general, there was a positive attitude towards the use of CDSS for breathlessness with various proposed features from the participants. Twelve high level workflow steps were identified with 5 as key points for intervention. Several proposed features such as reporting likelihood of causes of breathlessness in a patient, link with evidence-based recommendations, integration with clinical notes and patient education materials were translated into a prototype. Mixed-method studies are planned to assess its usability to inform subsequent iterations of the CDSS development.

Telehealth services for long-term monitoring of chronically ill patients are becoming more and more common, leading to huge amounts of data collected by patients and healthcare professionals each day. While most of these data are structured, some information, especially concerning the communication between the stakeholders, is typically stored as unstructured free-texts. This paper outlines the differences in analyzing free-texts from the heart failure telehealth network HerzMobil as compared to the diabetes telehealth network DiabMemory. A total of 3,739 free-text notes from HerzMobil and 228,109 notes from DiabMemory, both written in German, were analyzed. A pre-existing, regular expression based algorithm developed for heart failure free-texts was adapted to cover also the diabetes scenario. The resulting algorithm was validated with a subset of 200 notes that were annotated by three scientists, achieving an accuracy of 92.62%. When applying the algorithm to heart failure and diabetes texts, we found various similarities but also several differences concerning the content. As a consequence, specific requirements for the algorithm were identified.