Ebook: dHealth 2025

The innovation wave of recent years shows no signs of slowing down, as evidenced by ongoing advances in artificial intelligence (AI) and the transformative rise of large language models (LLMs), and the potential of these technologies to enhance healthcare has been widely demonstrated in recent literature. At the same time, these developments have been accompanied by new legislative efforts at a European level. For example, the European Health Data Space (EHDS) aims to establish a common framework for health data across the EU, the AI Act sets out regulations for the responsible development and deployment of AI applications, and the NIS2 Directive provides guidance on protecting critical infrastructure against emerging cyber threats. The successful transition of medicine and healthcare into the digital era will require a careful balance between these aspects.

Since 2007, the dHealth conference has provided an annual platform for scientists, healthcare professionals, stakeholders and decision makers to present, discuss and develop their ideas on digital health innovations, attracting around 300 participants each year. Staying true to its interdisciplinary mission, dHealth welcomes not only established experts from academia, industry, government, and healthcare organizations, but also students and young entrepreneurs with fresh ideas who might bring new perspectives and bold initiatives to the dHealth space.

This year’s 19th edition of the dHealth proceedings offers a diverse snapshot of ongoing research trends in health IT across Europe. The topics addressed at the 2025 conference placed a particularly strong emphasis on software, covering areas such as the adoption of emerging technologies like AI and LLMs, predictive analytics, virtual reality, interoperability, and standardization, while several papers also explored the ethical, legal, social, and economic implications of digital health. Together, these contributions form a valuable mosaic of the current state of eHealth and health IT research – one that may offer insights for the future transformation of medicine and healthcare systems.

Graz, Hall in Tyrol, Vienna, May 2025

Martin Baumgartner (AIT)

Dieter Hayn (AIT)

Bernhard Pfeifer (UMIT)

Günter Schreier (AIT)

Background:

Associative memory is essential for episodic memory formation, and training in this domain has been shown to enhance memory in both clinical and non-clinical populations. However, the application of VR-based training in this memory domain remains underexplored.

Objectives:

This study aimed to develop and evaluate a VR-based intervention to improve associative memory, focusing on verbal-visual and visual-auditory stimuli associations.

Methods:

Five healthy younger adults (mean age = 24.6) completed 8 trials of object-name and object-sound matching tasks in a virtual environment.

Results:

Significant performance improvements were observed across trials, with object-name matching showing higher recognition accuracy and faster response times than object-sound matching.

Conclusion:

Both tasks demonstrated increased accuracy with reduced response times with training. This study underscores the importance of tailoring VR-based cognitive training to specific associative tasks, offering promising applications in memory rehabilitation and cognitive enhancement.

Background:

This paper explores the viable design of collaborative game elements to promote physical activity in the context of digital health interventions.

Objectives:

We investigate the role of social relatedness in motivating users to engage in achieving step-count goals.

Methods:

The study utilised a minimalistic multiplayer step counter game called Shared Achievements, implementing group-based “collaborative effort” mechanics. Participants used the application with acquainted team members for two weeks.

Results:

Outcomes highlight the importance of communication and shared goals in fostering motivation and adherence. While group dynamics and social support enhanced engagement, challenges such as unequal contributions and competitive pressures were identified.

Conclusion:

The study underscores the need for careful design considerations to balance competition and collaboration, suggesting digital health tools can benefit from social achievement mechanics and incorporating motivational strategies tailored to different user preferences.

Background:

Europe’s ageing society faces a growing gap in long-term care (LTC) resources. The Care about Care project developed a Remote Care Assist (RCA) system to support LTC staff.

Objectives:

The project aimed to develop and evaluate a robust, mixed-reality- capable communication infrastructure for home care workers. This case report (a) details requirements for a communications infrastructure, (b) examines challenges in field testing, and (c) describes the transition to an alternative approach. It also provides a glimpse into a new solution used in a follow-up project.

Methods:

User requirements were gathered through co-creation workshops across three countries. Technical requirements focused on modularity, security, integration, and cross-platform usability. The initial implementation used Asterisk and was tested with ∼300 participants.

Results:

The field test revealed persistent video quality issues with Asterisk, attributed to its WebRTC implementation. A switch to Kamailio resolved these issues, improving video quality but sacrificing group call capabilities.

Conclusion:

Pre-testing is limited, highlighting the necessity of field testing. Emerging frameworks like LiveKit could be an alternative for follow-up projects.

Background:

Mobile health (mHealth) technology can support therapy adherence and self-management of people with multiple sclerosis (MS), but mHealth solutions tailored for the Austrian context are currently lacking.

Objectives:

This study aimed to evaluate usability, user experience and user acceptance of the Swiss “MS Active App” in an Austrian setting, and to identify transferable requirements and design implications.

Methods:

Nine people with MS used the MS Active App for one week as part of their individual physiotherapy and occupational therapy. Data collection included standardised questionnaires, a user diary, qualitative interviews with patients, and focus group discussions with therapists.

Results:

People with MS and their therapists rated the usability, user experience and user acceptance of the app as overall good. Qualitative accounts included a number of specific suggested improvements from which transferable requirements and design implications may be derived.

Conclusion:

The MS Active App offers a potentially suitable mHealth solution for people with MS in Austria, but several suggested improvements should be considered prior to implementation in practice.

Background:

Adherence to clinical guidelines supports high quality patient care. Conformance checking, a feature of process mining, can potentially automate the assessment of adherence to clinical guidelines in practice.

Objectives:

This paper investigates appropriate conformance checking in practice.

Methods:

Conformance checking in practice was simulated with generated test data, a FHIR server and process mining tools. A corresponding literature review was conducted in parallel.

Results:

Activities of clinical guidelines or in healthcare processes should be coded using clinical nomenclature to support conformance checking.

Conclusion:

SNOMED CT should be used as a nomenclature and activities should be coded with SNOMED concepts of the type “procedure”.

Background:

This thesis deals with the potentials and challenges of climbing for blind and visually impaired people.

Objectives:

The extent to which an innovative technical assistance system can contribute to supporting this sport in climbing gyms is analyzed.

Methods:

As part of this study, a needs analysis was carried out using a mixed methods approach. This comprises an observational study, expert interviews and a quantitative survey of the potential target group.

Results:

The studies show that climbing can promote the development of social and physical skills. Nevertheless, there are needs and challenges, such as finding the next hold on a route.

Conclusions:

The technical assistance system could provide support by addressing these specific challenges and needs of blind and visually impaired people when climbing and supporting them with acoustic signals.

Background:

Co-design workshops can challenge visualization skills.

Objectives:

To evaluate how micro design patterns support co-design workshops.

Methods:

In a workshop, participants designed low-fidelity prototypes for Patient-Reported Outcome Measures (PROM) visualizations using 12 pre-selected micro design patterns.

Results:

Patterns were rated as rather helpful; suggestions included better introduction and practical demonstrations.

Conclusion:

Micro design patterns can support co-design, but require some improvements in handling.

Background:

Cancer survivors often face long-term physical and psychosocial challenges post-treatment. Patient Reported Experience Measures (PREMs) are critical for understanding their perspectives and improving healthcare quality. However, no standardized tool currently exists for collecting these PREMs in Switzerland.

Objectives:

This study aimed to develop and test a digital application for collecting PREMs from cancer survivors to identify areas for improving aftercare.

Methods:

A web-application implementing the Cancer Patient Experiences Questionnaire (SCAPE-CH) using HL7 FHIR was developed, and usability was assessed using the System Usability Scale (SUS). Cancer survivors were recruited via social media to provide PREMs using the tool.

Results:

Data from 77 cancer survivors highlighted key areas for improvement, including communication between providers, information on long-term side effects and knowledge transfer between caregivers and doctors. SUS scores indicated good usability of the application.

Conclusion:

The digital PREMs tool effectively captured valuable insights, identifying critical areas for enhancing aftercare. Future work will focus on recruiting more patients to get representative data and integrating findings into broader healthcare strategies for cancer survivors.

Climate change is increasing acute heat events, intensifying health risks and straining healthcare systems. This study aims to support heat-related diagnoses prediction models for Germany by assigning ICD-10-GM codes to relevant conditions identified from the literature. Using the OHDSI mapping tool and clinical validation, 64 heat-related conditions were coded, enhancing data standardization. This approach facilitates reliable inclusion of diagnoses in association analyses and paves the way for improved resource allocation during heat events.

Background:

The acceptance and use of clinical decision support systems are often limited by insufficient contextual adaptation.

Objectives:

Identification of barriers to context assessment and requirements for an instrument to capture context factors.

Methods:

A questionnaire-based survey investigated requirements for a context assessment instrument.

Results:

Challenges in context assessment include insufficient knowledge of context factors and communication issues. Desired features are concise information and ease of use.

Conclusion:

An efficient instrument for capturing context factors is crucial to addressing gaps, improving communication, and fostering interdisciplinary collaboration in CDSS development.

Introduction:

Integrating nursing informatics education into nursing is essential for advancing the digital transformation of nursing. The objective was to develop nursing informatics curricula tailored to the needs in Kosovo and Israel.

Methodology:

Reviewing 16 international guidelines, conducting interviews with 22 national stakeholders, and undertaking an international Delphi study.

Results:

The study identified key challenges and recommendations for nursing informatics education. It ranked a list of 40 nursing informatics topics and developed two national curricula.

Discussion:

Effective nursing informatics education on a national scale requires the integration of national stakeholders and international perspectives to address diverse needs and contexts and to advance the field.

Introduction:

The rapid integration of digital health tools into clinical practice presents new opportunities for supporting clinical reasoning. But doctors and nurses must learn how they can use digital health tools such as AI and telehealth to support clinical reasoning (CR). Our objectives were to investigate how and which digital tools have been integrated in medical and nursing CR education.

Methodology:

Rapid review of 46 studies about integrating digital health tools in medical and nursing CR education.

Results:

Most studies reported about Large Language Models, EHRs and Clinical Decision Support Systems integrated into CR education.

Discussion:

Digital tools are increasingly integrated into medical and nursing clinical reasoning education, yet the focus is limited to certain types of tools.

Background:

Hospital-onset bacteremia (HOB) is one of the most common hospital-acquired infections, resulting in increased in-hospital morbidity and mortality. To support clinicians in assessing patients at risk, data-driven applications can be implemented as decision-support.

Objectives:

We aim at rigorously collecting clinician’s needs in an innovative, interdisciplinary tandem approach to understand requirements for a digital, in-hospital HOB risk assessment application with interactive visualization interfaces, incorporating users from the outset.

Methods:

Interviews and questionnaires for assessing (non-)functional requirements for data management and analysis (backend) and visualization (frontend) were developed. Thirteen professionals from eight German university hospitals participated.

Results:

For front- and backend, we identified requirements concerning data collection and provision, risk prediction, data input and export, and user interface.

Conclusion:

We successfully identified requirements classified into 4 groups, building a solid base for implementation of a user-centric, digital HOB risk assessment decision-support tool with interactive visualizations.

Background:

The accumulation of Real-World Data (RWD) from Electronic Health Records (EHRs) and registries offers substantial potential for generating Real-World Evidence (RWE). However, the ability to generate robust evidence from real-world data hinges on its quality. This is especially critical when heterogeneous data is first transformed into standardized, research-ready data models.

Objective:

This study presents an approach for assessing data completeness through a pipeline for extracting and transforming oncological RWD.

Methods:

We introduce a technical solution that enables the assessment of data completeness across three data transformation stages, beginning with the initial data source and extending through Health Level 7 (HL7) Fast Healthcare Interoperability Resources (FHIR) to CSV.

Results:

Using Trino, a distributed SQL engine, we evaluate data completeness at the three transformation stages by comparing cancer diagnosis counts. The modular pipeline design, compatible with various data sources, allows for error detection in ETL processes.

Conclusion:

Future work will expand the system to address additional data quality dimensions, such as correctness and plausibility, improving the overall robustness of data analytics in federated environments.

Background:

Effective infectious waste management in healthcare facilities is critical to public health and environmental safety.

Objectives:

This study aims to develop and implement a scalable digital dashboard with real-time analytics to enhance infectious waste management. It also evaluates staff knowledge, attitudes, and behaviors (KAB) while monitoring how these factors evolve over time following dashboard implementation.

Methods:

The ADDIE model guided the development of the dashboard. A pre- and post-intervention survey approach was used to assess changes in KAB. Data trends and waste generation rates were analyzed over time, with departmental breakdowns to identify areas for improvement.

Results:

Data from 29 staff members revealed that 79.3% demonstrated high knowledge, 82.76% had positive attitudes, and 89% exhibited effective behaviors. Strong correlations were found between knowledge and attitudes (r = 0.602, p < 0.001) and attitudes and behaviors (r = 0.673, p < 0.001). Over time, post-intervention surveys showed improvements in knowledge retention and behavioral compliance. The dashboard, integrated with IoT and BI analytics, provided real-time tracking, compliance alerts, and waste trend analysis. The study highlights the effectiveness of digital interventions in optimizing infectious waste management and ensuring regulatory compliance.

Classic rehabilitation faces different challenges, such as low patient engagement, limited motivation, or difficulties in personalizing the therapy, which is needed for good recovery outcomes. Serious games present a promising solution to these problems by enhancing the individual motivation and deliver interactive rehabilitation possibilities. To investigate their potential, a preliminary study was conducted using an online questionnaire as the methodological approach. Overall, 15 therapists (average age: 33.1 years; average experience: 9.7 years) from various professional backgrounds participated in it. The results showed that seven therapists (47%) already used serious games within patient rehabilitation, while eight (53%) had not yet used them. Most of the already used solutions also only cover cognitive rehabilitation and do not use any additional sensors to capture movements. Only within 2 cases serious games were also used at home in addition to the therapeutical setting. In addition, the participants also evaluated five serious games integrated into a rehabilitation platform, highlighting strengths such as increased patient motivation and adaptability, but also missing features like tailored feedback systems and the need for improved integration into their daily workflows. According to the discussion, these findings suggest that serious games in rehabilitation are not yet broadly in real use. There is also the need for better accessibility, functionality, time and therapist training to optimize the use of serious games and address some current limitations in rehabilitation practices.

Background:

The ActiveWaiting App was designed to support interspersed bouts of health-promoting physical activity (PA) during waiting periods and other idle time.

Objectives:

The aim of this study was to pilot a study design to investigate the potential impact of the ActiveWaiting App on PA behavior and health-related quality of life.

Method:

A randomized waitlist-control design with one-week intervention and control phases was applied. Self-reported PA was recorded twice daily. Quality of life was measured using the EuroQol questionnaire.

Results:

Thirty-three adults (age 28-61 years) who held sedentary jobs were recruited and used the app on average (median) once (range 0-22 times) during the one-week intervention phase. Preliminary analyses indicate a trend towards increased PA during intervention compared to control phases, and small increases in quality of life during all phases.

Conclusion:

The ActiveWaiting App shows potential as a low-threshold intervention to promote PA. Future studies need to address data completeness, specifically of PA measurements.

An increasing number of countries has been introducing nationwide Electronic Health Records (EHRs) with the aim of improving healthcare processes and enhancing accessibility for patients and caregivers. To ensure these systems effectively meet users’ needs, the continuous process of designing and developing EHRs has to be based on a user-centered approach, focusing on the everyday tasks and challenges of users while interacting with the medical system. This study specifically examines the needs of intra-familial, informal caregivers in the context of nationwide EHRs. Based on an online survey with 608 potential users of EHRs for intrafamilial care, the essential key features have been identified, analyzed, and prioritized. The findings highlight the significance of functionalities supporting caregivers in daily tasks. This research delivers initial insights on aspects regarding potentials and barriers of EHRs for informal care in a familial context.

Background:

Type 2 diabetes (T2D) continues to present a global public health challenge due to its increasing prevalence. Early diagnosis is critical for preventing complications, but current screening methods often fail to detect early diabetic conditions.

Objectives:

This study aimed to classify T2D patients from healthy individuals using high-resolution N-glycan profiling.

Methods:

Glycan profiling was performed on serum samples from 161 individuals using capillary electrophoresis with laser-induced fluorescence detection. Different classification methods were fine-tuned using hyperparameter optimization and feature selection techniques, and their performance was comprehensively evaluated based on quality metrics.

Results:

The Extra Trees Classifier outperformed the other models with the highest median AUC, demonstrating robust accuracy (0.8982), sensitivity (0.8966), and specificity (0.9000).

Conclusion:

N-glycan profiling combined with machine learning provides a promising approach for early T2D detection. The Extra Trees Classifier showed exceptional predictive performance, warranting further investigation with larger datasets to validate its clinical applicability.

Background:

This work explores the interaction design around the approach of making use of otherwise idle waiting times to perform short bouts of exercise.

Objectives:

We provide insights on (1) how digital tools can support the utilisation of waiting times for physical activity, and (2) how an application that makes exercises suggestions in a manner designed to be frictionless is experienced by users.

Methods:

We developed the Active Waiting (AW) app, a prototype for addressing physical activity barriers such as time and effort, as well as an additional barrier related to the social awkwardness linked to performing exercises in public. AW was iteratively evaluated through guerrilla testing and a field study.

Results:

Study participants appreciated the application for making waiting times more interesting, offering viable exercise suggestions, encouraging the taking of additional (active) breaks, and most of all the possibility to select unobtrusive exercises.

Conclusion:

The study indicates that unobtrusive and easily accessible exercises integrated into waiting times can effectively reduce barriers to physical activity. While the app was well-received for enabling meaningful use of waiting time, future work could explore more proactive approaches to encourage consistent usage in diverse daily contexts.

Postoperative pain is a relevant and unresolved problem in clinical practice. In order to reduce the occurrence of severe postoperative pain, preventive, multi-professional and target group-specific pain management should be implemented. Risk assessment models based on machine learning and artificial intelligence are a resource-efficient way to identify the target group. The aim of this study was to develop a risk assessment model for early predicting poor postoperative pain outcomes that achieves good results without the need of additional, non-routine data collection. The various machine learning-based models were developed by using electronic medical records from over 70.000 in- and outpatient cases and 807 modelling features. The GBM (gradient boost machine) algorithm performed best with an area under the receiver operating characteristic curve (AUROC) of 0.82 on hold-out test data. Despite the excellent result, further research is needed to determine the modelt’s performance in clinical practice.

This paper examines the challenges and presents preliminary ideas on how Personal Health Data Spaces (PHDS) in Decentralized Content-Addressable Storage (DCAS) networks can serve as both an alternative to and a vital contribution towards the concept of a Health Record Bank (HRB) within the framework of the European Health Data Space (EHDS).

With the digitalization of the healthcare system the need for standardization of data access has drastically increased. At the same time a push to more strongly include patients in the medical process created demand for them to have more access on their own data. These developments led to the creation of both FHIR and SMART on FHIR and has peaked in the Health Outcomes Observatory (H2O) project’s effort to establish patient-reported outcomes as a standard tool for patient perspective inclusion in healthcare. This paper aims to show that these technologies allow to achieve the goals set out by the H2O project using solely open-source solutions.

Background:

Perioperative blood pressure data often contain artifacts that can compromise data integrity for clinical decisions and research.

Objectives:

The main objective of this retrospective analysis was to evaluate the efficiency and reliability of various algorithms for artifact detection in perioperative blood pressure data, specifically assessing their performance in different clinical scenarios and measurement methods.

Methods:

Data from 106 patients at the Medical University of Vienna were analysed using algorithms based on the interquartile range, Z-Score, Cut-off methods, and Moving Mean/Median. Validation involved comparisons against a reviewer standard set by anaesthesia experts.

Results:

Using a standard deviation based algorithm was most effective, offering superior accuracy and reliability across scenarios although sensitivity was below 60% for all used algorithms.

Conclusion:

Our results support a scenario-specific approach to artifact detection, underlining the need for research into adaptive algorithms that enhance data quality for clinical and research applications.