Ebook: dHealth 2023

This book presents the proceedings of the 17th annual conference on Health Informatics meets Digital Health (dHealth 2023), held in Vienna, Austria, on 16 and 17 May 2023. The conference series provides a platform for researchers and decision makers, health professionals, healthcare providers, and government and industry representatives, giving them a forum in which to present and discuss innovative digital health solutions with the aim of improving the quality and efficiency of healthcare using digital technologies. The d in dHealth stands for a collection of important topics in the area of medical informatics, encompassing concepts such as digitalisation, datafication and data-driven decision making, as well as predictive modelling and “deep” health for better patient outcomes and sustainability in healthcare.

The conference also supports student attendees, providing an opportunity for them to complement their education in the respective tertiary teaching schools and the chance to act as presenters in the student competition. It is crucial in engaging the next generation to become digital-health developers, architects, facilitators, pioneers, innovators, and managers.

The papers give an insight into state-of-the-art aspects of dHealth, including design and evaluation of user interfaces, patient-centred solutions, electronic health/medical/patient records, telemedical approaches and solutions, predictive models, machine learning in healthcare and biomedical data analytics, and the book offers the reader an interdisciplinary overview of current research activities in digital health.

Bernhard Pfeifer (UMIT)

Günter Schreier (AIT)

Martin Baumgartner (AIT)

Dieter Hayn (AIT)

Graz, Hall in Tyrol, Vienna, May 2023

Background:

To deploy clinical decision support (CDS) systems in routine patient care they have to be certified as a medical device. The European Medical Device Regulation explicitly asks for the use of standards and interoperability in the approval process.

Objectives:

We extended an existing dermatological CDS system with emerging standards for CDS interoperability, to facilitate a future integration into existing healthcare infrastructure, and approval as a medical device.

Methods:

The data collection part of a CDS system was extended with the endpoints required by the CDS Hooks specification. FHIR QuestionnaireResponse resources trigger a newly defined hook.

Results:

One hundred and seventeen clinical observations and patient variables needed for the ranking of a disease were mapped to SNOMED CT or LOINC and modeled as FHIR Questionnaire which is rendered using LHC LForms in a SMART on FHIR app with the SMART Dev Sandbox.

Conclusion:

SMART on FHIR in combination with CDS Hooks facilitates the integration of existing CDS systems into EHR systems, potentially improving education and patient care.

Background:

The need for software suppliers to react swiftly to the plethora of application requests and constantly shifting market requirements is one of the major problems facing the health IT business in the context of digital health transformation. This can only be achieved when the necessary staff and resources are available.

Objectives:

The objective of this work is to identify challenges health IT companies are confronted with related to personnel capacities and skilled workers.

Methods:

Using a questionnaire distributed through newsletters and social media among representatives of software companies and hospitals we collected information on current hurdles of health software providers and their strategies to overcome these in order to address the demands of the digital health transformation.

Results:

The main findings of the survey are that scarce resources in software development are among the reasons for not achieving strategic goals on time in the health IT sector and for not being able to react flexibly to market changes. A strategy to overcome missing expert knowledge and own resources without free capacity is to hire external resources.

Conclusions:

With the ever-changing landscape of digital health, it is essential to have skilled workers with knowledge on the peculiarities of clinical workflows. The existing shortage of skilled workers leads to a reduction of innovative power in the health IT sector, potentially slowing down the digital health transformation.

Background:

Current monitoring and evaluation methods challenge the healthcare system. Specifically for the use case of immunization coverage calculation, person-level data retrieval is required instead of inaccurate aggregation methods. The Clinical Quality Language (CQL) by HL7^®, has the potential to overcome current challenges by offering an automated generation of quality reports on top of an HL7^® FHIR^® repository.

Objectives:

This paper provides a method to author and evaluate an electronic health quality measure as demonstrated by a proof-of-concept on immunization coverage calculation.

Methods:

Five artifact types were identified to transform unstructured input into CQL, to define the terminology, to create test data, and to evaluate the new quality measures.

Results:

CQL logic and FHIR^® test data were created and evaluated by using the different approaches of manual evaluation, unit testing in the HAPI FHIR project, as well as showcasing the functionality with a developed user interface for immunization coverage analysis.

Conclusion:

Simple, powerful, and transparent evaluations on a small population can be achieved with existing open-source tools, by applying CQL logic to FHIR^®.

The rehabilitation process after knee injuries is often challenging for patients and requires a high level of resilience, as it involves the frequent repetition of mostly monotonous exercises. Based on recent research, serious games can significantly improve motivation by merging exercising with entertainment aspects and even combining it with hardware to apply external tasks and track the progress. The aim of this research is to propose and evaluate a new serious game pattern. The development is performed using systematic feedback from domain experts. The test setup involves analysis of patients’ feedback. The final game comprises an interaction with a balance board and an attached smartphone. Evaluation showed two main results. From a technical point of view: sensors of a standard smartphone (and it’s sensitivity) paired with a PC and its screen are usable in a rehabilitation setting. From a psychological point of view: the motivation to perform the knee rehabilitation process can be enhanced with a serious game delivering entertaining aspects to it.

Background:

Artificial Intelligence (AI) has had an important impact on many industries as well as the field of medical diagnostics. In healthcare, AI techniques such as case-based reasoning and data driven machine learning (ML) algorithms have been used to support decision-making processes for complex tasks. This is used to assist medical professionals in making clinical decisions. A way of supporting clinicians is providing predicted prognoses of various ML models.

Objectives:

Training an ML model based on the data of a hospital and using it on another hospital have some challenges.

Methods:

In this research, we applied data analysis to discover required data filters on a hospital’s EHR data for training a model for another hospital.

Results:

We applied experiments on real-world data of ELGA (Austrian health record system) and KAGes (a public healthcare provider of 20+ hospitals in Austria). In this scenario, we train the prediction model for ELGA- authorized health service providers using the KAGes data since we do not have access to the complete ELGA data.

Conclusion:

Finally, we observed that filtering the data with both feature and value selection increases the classification performance of the prediction model, which is trained for another system.

Background:

Mobile health (mHealth) apps are increasingly used in healthcare to support people with chronic diseases such as diabetes. mHealth acceptance is crucial for using them. Due to acceptance problems, however, mHealth apps are not used by all chronic disease patients. To predict user acceptance, technology acceptance models such as UTAUT2 are used. However, UTAUT2 was not explicitly developed for the mHealth context.

Objectives:

This study investigates if additional health-related constructs could increase the predictive power of the UTAUT2 model.

Methods:

A mixed-methods design, comprising an initial qualitative methods triangulation study that consisted of a literature search, expert interviews, and patient interviews, and a subsequent quantitative cross-sectional survey with 413 patients was used.

Results:

The mixed-methods study revealed and validated two new constructs relevant for predicting mHealth acceptance not represented in the UTAUT2 model: “perceived disease threat” and “trust”.

Conclusion:

The UTAUT2 model was successfully extended by two new constructs relevant to the mHealth context.

Background:

Blood collection centers can take advantage of the huge amount of data collected on donors over the years to predict and detect early the onset of several diseases, However, dedicated tools are needed to carry out these analyses.

Objectives:

This work develops a tool that combines available data with predictive tools to provide alerts to physicians and enable them to effectively visualize the history of critical donors in terms of the parameters that led to the alert.

Methods:

The developed tool consists of data exchanging functions, interfaces to raise alerts and visualize donor history, and predictive algorithms. It was designed to be simple, modular and flexible.

Results:

A prototype was applied to the Milan department of the Associazione Volontari Italiani Sangue, and was deemed suitable for prevention and early diagnosis objectives by the physicians of the center. The included Machine Learning predictive algorithms provided good estimates for the variables considered in the prototype.

Conclusion:

Prevention and early diagnosis activities in blood collection centers can be effectively supported by properly using and displaying donor clinical data through a dedicated software tool.

Background:

Long-term care faces severe challenges on the supply (shortages of formal and informal carers) as well as on the demand side (increasing number of care-dependent people). To cope with these challenges, new forms of support for the professional care network are needed.

Objectives:

This paper describes the concept and implementation of a Remote Care Assist (RCA) service, consisting of a web-application for the Care Expert Center (CXC) and Remote Support (RS) applications for the HoloLens 2 as well as for Android and iOS smartphones.

Methods:

Using the evidence-based and user-centred innovation process (EUIP), a Remote Care Assist service was conceptualized and implemented for home care service settings in three European countries.

Results:

After five iterations within two phases of the EUIP, the final feature set of the RCA-service was determined and implemented.

Conclusion:

By working closely with the target group, it was possible to identify potential hurdles and additional requirements such as a well-thought-out interaction concept for the HoloLens or a good organizational embedding of the service.

In order to perform in vitro cardiotoxicity screening of pharmacological substances, multi-electrode array systems are increasingly used to measure the extracellular field potentials of cell layers of human induced pluripotent stem cell cardiomyocytes. The analysis of the field potentials is usually performed using complex analysis software provided by the hardware manufacturers. In the case of the Cardiac Analysis Tool software from Axion Biosystems, inconsistencies were found in the results, which can significantly influence the cardiotoxicity screening results. In order to obtain more reliable results, a new algorithm was developed and implemented in an easy-to-use software tool, the INCardio Data Analysis Tool, which, due to its high degree of automation, can also be used by inexperienced users. The validation reveals differences in the results of the two tools both in depolarization spike amplitudes and in the time course of the field potential durations. The manual analysis of all signals affected by deviations shows that the results of the newly developed Data Analysis Tool are correct in all cases and can therefore be classified as more accurate and reliable than the reference analysis software.

The Austrian nationwide EHR system ELGA can contribute valuable data for research due to its high volume of data and broad population coverage. In order to be applicable in international research projects, transformation to a standardized, research-oriented data model such as the OMOP common data model is essential. In this paper we describe our experience with the corresponding transformation task. Using Python scripts, we implemented a prototypical process that extracts, transforms, maps, and loads fully structured sections of ELGA documents to an OMOP database.

Radiologists rarely interact with the patients whose radiological images they are reviewing due to time and resource constraints. However, relevant information about the patient’s medical history could improve reporting performance and quality. In this work, our objective was to collect requirements for a digital medical interview assistant (DMIA) that collects the medical history from patients by means of a conversational agent and structures as well as provides the collected data to radiologists. Requirements were gathered based on a narrative literature review, a patient questionnaire and input from a radiologist. Based on these results, a system architecture for the DMIA was developed. 37 functional and 17 non-functional requirements were identified. The resulting architecture comprises five components, namely Chatbot, Natural language processing (NLP), Administration, Content Definition and Workflow Engine. To be able to quickly adapt the chatbot content according to the information needs of a specific radiological examination, there is a need for developing a sustainable process for the content generation that considers standardized data modelling as well as rewording of clinical language into consumer health vocabulary understandable to a diverse patient user group.

Fatigue is the most prevalent Long-COVID symptom. Individuals who are affected have to learn to organize and manage daily activities according to the subjectively perceived energy reserves. Our objective was to develop an application, Fading Fatigue, that supports patients in their energy management, in particular after an initial therapy guided by health professionals. Fading Fatigue was developed in an iterative approach and implemented as a client-server application. Interviews and a literature search were conducted to identify limitations and challenges of the current treatment. Fading Fatigue offers several tools for energy management: a daily energy planner, a documentation aid for well-being and a progress view. Future work should study usability. Inclusion of additional features increasing the adherence such as providing feedback could be considered.

In this paper we outline a cost-effective design for the improvement of telemedicine applications through a two components concept. A normal smartphone is equipped with a small device containing different filter options, which provide particular spectral information for the identification of skin lesions. By merging the measured spectra, a higher data density and more information can be obtained using an ANN to improve an early diagnosis of skin lesions through telemedicine applications in remote areas.

Background:

Quality indicators (QI) are a common method to ensure quality in healthcare. This paper is based on the so-called QI-KA project, which defined cross-sector QI for the Austrian healthcare system. However, to allow for automated conformance checking, the QI must be modelled in a formal way.

Objectives:

The aim of this paper is to gather requirements on modelling languages and tools in healthcare, create models for one of the QI-KA project’s QI and finally evaluate them.

Methods:

The QI-2 is modelled in the Business Process Model and Notation (BPMN) together with the Decision Model and Notation (DMN) to showcase and evaluate their application and suitability.

Results:

The generated models show that BPMN and DMN are mostly appropriate for this use case and serve as a basis for automated conformance checking.

Conclusion:

We successfully showed that BPMN and DMN can be used to model cross-sector QI in a formal way to prepare for conformance checking. The field of application can be extended to other medical areas. To further improve quality in healthcare, outcomes from models and conformance-checking should be discussed in interdisciplinary teams.

Background:

Aphasia describes the lack of the already gained ability to use language in a common way. “Language” here covers all variations of forming or understanding messages.

Objectives:

The APH-Alarm project aims to develop a service concept that provides alternative communication options for people with Aphasia to trigger timely help when needed. It considers that a typical user may not be familiar with modern technologies and offers several simple and intuitive options.

Methods:

The approach is based on event detection of gestures (during daytime or in bed), movement pattern recognition in bed, and an easy-to-use pictogram-based smartphone app.

Results:

Agile evaluation of the smartphone app showed a promising outcome.

Conclusion:

The idea of a versatile and comprehensive solution for aphasic people to easily contact private or public helpers based on their actions or automatic detection is promising and will be further investigated in an upcoming field trial.

Background:

Rehabilitation plays a key role in the recovery of upper extremity function after breast cancer surgery. Motion capture (mocap) systems for serious gaming have shown the potential to enable home-based rehabilitation, but clinical accuracy needs to be examined.

Objectives:

Validation of markerless mocap systems for telerehabilitation after breast cancer surgical intervention.

Methods:

The accuracy of the markerless mocap device Azure Kinect in detecting compensatory movements and postural disturbances has been compared to a gold standard Optitrack system in five volunteers. Subsequently, a serious game for mocap-based shoulder exercises has been developed and integrated into a telerehabilitation platform.

Results:

The Azure Kinect shows good reliability for scapular elevation (ICC >0.80; MAE <2.1 cm) and trunk tilt (ICC=0.88; MAE=5°), moderate reliability for rounded shoulders (ICC=0.51; MAE=2.6cm) and poor reliability for kyphosis angle (ICC=0.22; MAE=18°).

Conclusion:

The Azure Kinect provides reasonable performance for shoulder rehabilitation. The proposed telerehabilitation platform has been tested by rehabilitation specialists and received positive feedback.

The COVID-19 pandemic brought forth rapid responses and changes in the acceptance of digital health interventions. Digital solutions appear increasingly promising, yet little is known about the peculiarities in the psychiatric context, contrary to other medical branches. The project MeHealth aimed at disclosing specific needs and reservations of patients and professionals in the psychiatric field. Apprehensions towards technology were found to be held on both sides. Cooperating with a psychiatric hospital in Austria, through a transdisciplinary research approach including focus groups and workshops, a framework for an integrated Digital Mental Health Tool was established. The findings leading to the framework show a strong need for patient-empowerment, enhancement of trust in technology and the need for multi-stakeholder cooperation. Digital tools should be designed to enhance the continuity of care and information exchange on behalf of the patient. Learnings were gained, which prove recommendable for future R&D projects on digitalization in the delicate field of psychiatry.

Equipping rooms used for medical purposes, like e.g., intensive care units, is an expensive and time-consuming task. In order to avoid extensive subsequent adjustments due to inappropriate layout visualization or geometric conditions difficult to identify in 2D plans, it is of utmost importance to provide an optimal planning environment to future users such as physicians and nurses. In this paper we present the concept of a fully automatized pipeline, which is designed to visualize computer aided design (CAD) data using virtual reality (VR). The immersive VR experience results in improvement of efficiency in the decision-making process during the planning phase due to better spatial imagination. The pipeline was successfully tested with CAD data from existing Intensive Care Units. The results indicate that the pipeline can be a valuable tool in the field of spatial planning in healthcare, due to simple usage and fast conversion of CAD data. The next step will be the development of a plugin for CAD tools to allow for interactions with the CAD models in Virtual Reality, which is not yet possible without manual intervention.

Background:

In the federal state of Salzburg, the harmonization of nursing internship took place from 2014–2019 as a joint effort of educational and internship providers. Currently the handling of these mostly paper pencil-based documents involves a manual process through different institutions and people.

Objectives:

The project provides the basis for the implementation of the “internship platform”. In the future, this is to be the digital, state-of-the-art one-stop shop for the state-wide practical nursing training (in the form of immanent internships) at all training levels.

Methods:

The process is influenced by modern requirements engineering techniques: As-is analysis of the internship process and related documents, contextual inquires in different internship providers, iterative focus group discussions focusing first on user stories, then on interface designs, and final user testing.

Results:

Standardized workflow and authorization concept for all user groups, mandatory requirements for the software system, tested user interfaces, tender documents for EU-wide tender procedure.

Conclusion:

Positive feedback from all involved user groups on project goal, results and involvement in the process.

Background:

Relearning physiological movement patterns is a key factor to success in the treatment of functional deficits. Motivation to train sustainably is essential for successful motor re-education and can be promoted by instrumentally supported real-time feedback.

Objectives:

Study findings should improve the understanding of real-time feedback visualization for exercises targeting the lower extremities.

Methods:

A mixed-methods survey on recognition, comprehensibility, color scheme and shape of six real-time feedback prototype visualizations was conducted among three user groups (physicians, physiotherapists, and patients).

Results:

The mean correct recognition of body regions visualized in the feedback was 55 %, ranging from 29 % to 74 %. Comprehensibility, color scheme and shape were best received for feedback with clear visual guidance, sympathetic and motivating color schemes and abstract visualizations of body regions.

Conclusion:

Insights were gathered for the design, optimization, and customization of visualizations to develop a real-time feedback prototype.

Background:

In emergency trauma room, adequate preparation of all resources prior to the patient’s arrival is essential to ensure optimal continuation of the treatment. Therefore, a good transfer of information between pre-hospital and hospital is very important, for example through networking technologies.

Objectives:

The aim is to identify what pre-hospital information is needed to ensure that all necessary resources in the ETR are optimally prepared for the incoming trauma patient.

Methods:

A qualitative, semi structured interview was conducted with physicians of ETR team at four trauma centers.

Results:

Physicians mentioned similar requests for pre-hospital information. The workflow in ETRs differed in alerting of team members and transferring of pre-notification information.

Conclusion:

Clinical needs for pre-hospital information for future development of support systems in the networking of accident site and hospital could be identified.

The JITAI is an intervention design to support health behavior change. We designed a multi-level modeling framework for JITAIs and developed a proof-of-concept prototype (POC). This study aimed at investigating the usability of the POC by conducting two usability tests with students. We assessed the usability and the students’ workload and success in completing tasks. In the second usability test, however, they faced difficulties in completing the tasks. We will work on hiding the complexity of the framework as well as improving the frontend and the instructions.

Stress is an increasing burden for our society and related to cardiovascular (CV) parameters and diseases. Effects of mental or physical stress were observed in CV parameters during task completion and recovery. These effects were measured using a novel handheld device, which can be incorporated in mHealth solutions.

Background:

There are many medical knowledge bases with potential for supporting medical professionals in their decision-making during routine care, yet usage of these sources remains low. Standardized linking of Clinical Decision Support (CDS) applications and existing medical knowledge bases is not a common practice.

Objectives:

Using existing eHealth standards to increase the utilization of knowledge bases and implement a prototype.

Methods:

Linking an existing online knowledge base via a FHIR CodeSystem supplement to the Austrian national EHR (ELGA) terminology server and accessing these data using CDS Hooks and FHIR.

Results:

We tested the approach by incorporating photosensitivity data of medications into a local copy of the Austrian terminology server. These data are directly used by a CDS Hooks compliant CDS service.

Conclusion:

The Austrian Terminology Server could be an important interface to access existing knowledge bases from within EHR systems. FHIR and CDS Hooks could lead the way for a simple and open integration of CDS services into EHR systems.