Ebook: Healthcare of the Future 2022

Healthcare of the Future 2022

Digital health – From vision to best practice!

International Conference Biel/Bienne

20 May 2022

We are in the aftermath of a pandemic that triggered considerable advances in the digital tools supporting healthcare. Contact tracing apps supported the work of healthcare professionals while other apps collected symptoms to learn more about the disease and its development. Modern IT-based solutions scanning the web with bot technology for news about COVID cases, such as the COVID-19 Dashboard of the Johns Hopkins University,(https://gisanddata.maps.arcgis.com/apps/dashboards/bda7594740fd40299423467b48e9ecf6.) initially delivered faster and better data than many national health authorities, who underestimated the efforts for collecting such data at the beginning of the pandemic.

We have also witnessed public authorities that were overwhelmed by the piles of infection reports sent on paper from GPs and hospitals using ordinary fax machines; authorities sometimes estimated the number of newly infected patients by weighing the incoming paper reports on a set of digital scales to deal with the data.

Each of us has attended many digital meetings using digital tools, and we were lucky that these tools were able to scale up to the sudden need and enable some continuity in communication when no one was allowed to travel or even leave the house. But we also experienced the fact that the tools can only partially replace personal contact and face-to-face discussions, which were not possible for considerable time spans. Now, in the transition back to normality, we are glad to invite you to the 2nd conference on the “Healthcare of the Future”. Our goal is to enable networking and exchange among researchers, thus we had to delay the conference for a year due to the pandemic. We will be one of the first medical informatics conferences in 2022 to once again permit personal attendance.

In the first edition of “Healthcare of the Future” in 2019, we tried to forecast digitised healthcare in the year 2030. We foresaw a further increased life expectancy (now to be questioned) and an improved, self-determined life at home with the use of intelligent systems, wearable devices and telemedicine services. We defined digital workflows for our elderly Swiss lady, Elisabeth Brönnimann-Bertholet, who suffers from diabetes and hypertension. An integrated cross-institutional clinical pathway was drawn up for her progredient hip arthrosis, which required surgery and a total hip endoprosthesis. We imagined a link between the patient at home, the GP, the specialist, the hospital and the rehabilitation centre by digital means.

In the pandemic however, we were forced to the realisation that the mix of infectious and healthy patients alone can be enough to put a modern hospital completely out of operation [1]. On the other hand, we also observed the use of modern robot technology for telepresence and physical distancing. It was even demonstrated that pre-diagnosis in public places using temperature checks and questionnaires could be performed by robots [2].

In 2019, we talked about personalised medicine. Now we have witnessed the development and worldwide availability of efficient vaccinations based on mRNA technology in record time, including the necessary clinical trials [3].

The first edition of the conference was held as part of the multi-stakeholder research project “Hospital of the future live” which had the goal of defining use cases for cross-sectoral treatment [4]. In its 2nd edition, we are just at the start of a novel competence centre ´nHospital@Home˙z at the Institute for Medical Informatics of the BFH. This competence centre aims to examine if and what kinds of current inpatient treatment could be realised in the patient’s home, and whether it is possible to integrate outpatient services to realise an improved care process. Thus, interoperability between the professional stakeholders in healthcare, and increasingly also with the patient, is an emerging topic.

This is the setting for the 2022 edition of “Healthcare of the Future”. The first conference had the goal of turning parts of the described visionary scenario into a tangible reality, with examples realised in our laboratory environment. In 2022, we will see which visions of the “Hospital of the future” have turned into reality and how digital interaction between nurses, caregivers, patients and healthcare institutions is to be realised.

The ultimate goal remains the same: improving and accelerating healthcare processes. Our three keynotes well reflect this by addressing key topics in medical informatics: interoperability, data quality and artificial intelligence (AI) with its ethical implications.

∙ “Transforming the healthcare ecosystem towards better interoperability” by Silvia Thun.

∙ “Data quality – What do we want, what do we need, what can we pay for?” by Rainer Röhrig

∙ “Ethical implications of AI usage in healthcare” by Tanja Krones

The conference comprises two tracks: main track and young researchers’ track. The four scientific sessions deal with the topics:

∙ Advancing Interoperability

∙ Semantic Interoperability

∙ Medical Informatics for Medical Research

∙ Evaluation of IT Influence

The young researchers’ sessions will focus on apps to support patients and healthcare professionals, applications in research in medicine and medical informatics and methods and strategies for workflow-based support in patient care.

Biel/Bienne April 4th 2022

Thomas Bürkle, Kerstin Denecke, Jürgen Holm, Murat Sariyar and Michael Lehmann

Bern University of Applied Sciences, Biel, Switzerland

References

[1] M. Nacoti, A. Ciocca, A. Giupponi, P. Brambillascam F. Lussana, M. Pisano, G. Goisis, D. Bonacina, F. Fazzi, R. Naspro, L. Lonfghi, M. Cereda, C. Montaguti, At the Epicenter of the Covid-19 Pandemic and Humanitarian Crises in Italy: Changing Perspectives on Preparation and Mitigation. NEJM catalyst innovations in Care delivery March (2020) 1–5. DOI: 10:1056

[2] L. Aymerich-Franch, I. Ferrer (2020). The implementation of social robots during the COVID-19 pandemic. ArXiv preprint. ArXiv:2007.03941

[3] E. Checcucci, F. Piramide, A. Pecoraro, D. Amparore, R. Campi, C. Fiori, O. Elhage, P. Kotecha, A. Vyakarnam, S. Serni, P. Dasgupta, F. Porpiglia, The vaccine journey for COVID-19: a comprehensive systematic review of current clinical trials in humans. Panminerva Med. 2022 Mar;64(1):72–79

[4] T. Bürkle, K. Denecke, M. Lehmann, E. Zetz, J. Holm, Integrated Care Processes Designed for the Future Healthcare System. Stud Health Technol Inform 245 (2017), 20–24.

Mobile apps indicate a positive effect on suicidal ideation and potential impact on suicide attempts. As part of the SERO suicide prevention program, Lucerne Psychiatry in collaboration with partner organizations aims to reduce suicides and suicide attempts in its service area, and to improve the self-management of suicidal individuals with a mobile app. The concept for such an app was developed in a trialog with health professionals, persons at risk and their relatives and its functions were compared to six known essential app-based strategies for suicide prevention, such as the development of a safety plan, access to support networks and tracking of mood. We present the concept and architecture for the app and discuss potential added value, which may result from the intertwining of the strategies within the app, which will be available in its first version in late 2022.

Healthcare processes have many particularities captured and described within standards for medical information exchange such as HL7 FHIR. BPMN is a widely used standard to create readily understandable processes models. We show an approach to integrate both these standards via an automated transformation mechanism. This will allow us to use the various tools available for BPMN to visualize and automate processes in the healthcare domain. In the future we plan to extend this approach to enable mining and analyzing executed processes.

Language barriers hamper or delay delivery of urgent and emergency care to migrant children when they or their parents don’t speak any of the languages commonly spoken in Switzerland. In such situations, nurses often fall back to use ad hoc communication aids, including translation apps and visual dictionaries, to collect information about a patient’s medical history. In this paper, we report on the participatory design process for a novel image-based communication aid. It is specifically tailored to the needs of migrant patients and nurses within Swiss pediatric clinics. We collected requirements in surveys and in-depth interviews with pediatric nurses. A prototype app was developed and tested with users in a scenario-based usability test. The results clearly show that the images developed, especially for symptoms, accidents or nutrition and excretion, are well comprehensible for triage and anamnesis. In contrast, a temporal classification or chronological occurrence of health incidents is difficulty to express with images.

Among medical applications of natural language processing (NLP), word sense disambiguation (WSD) estimates alternative meanings from text around homonyms. Recently developed NLP methods include word vectors that combine easy computability with nuanced semantic representations. Here we explore the utility of simple linear WSD classifiers based on aggregating word vectors from a modern biomedical NLP library in homonym contexts. We evaluated eight WSD tasks that consider literature abstracts as textual contexts. Discriminative performance was measured in held-out annotations as the median area under sensitivity-specificity curves (AUC) across tasks and 200 bootstrap repetitions. We find that classifiers trained on domain-specific vectors outperformed those from a general language model by 4.0 percentage points, and that a preprocessing step of filtering stopwords and punctuation marks enhanced discrimination by another 0.7 points. The best models achieved a median AUC of 0.992 (interquartile range 0.975 – 0.998). These improvements suggest that more advanced WSD methods might also benefit from leveraging domain-specific vectors derived from large biomedical corpora.

Many patient portals have been introduced and evaluated in recent years. The results of evaluation studies are difficult to compare, however, as the evaluated patient portal is often not clearly or only incompletely described in the publication. This problem is common to evaluations in health informatics. We evaluated the completeness of descriptions of patient portals in 15 exemplary evaluation publications using the TOPCOP taxonomy. Our results show that core functionalities such as portal design, patient communication, educational features, or system notifications were quite clearly described in all 15 evaluation studies. Other descriptions, such as web accessibility or data management, were often not provided. We conclude that taxonomies such as TOPCOP should be used and even required for describing interventions in evaluation papers.

ABIDE_MI is a complementary funded 18 months project within the German Medical Informatics Initiative (MII), which aims to align IT infrastructures and regulatory/governance structures between biobanks/biobanking IT and the MII data integration centres (DIC) at German university hospitals. A major task in 2021 was the systematic collection of all documents describing rules, as well as proposal/contract templates for data and biosample use and access at each of the participating 24 university hospitals and their comparison with MII-wide consented data sharing principles, documents and governance structures. This comparison revealed large heterogeneity across the ABIDE_MI sites and further, redundant structures/regulations currently established at the German university hospitals. A second task was the design and stepwise development of an IT network infrastructure with central components (data and biosample query portal) and decentralized standardized FHIR servers to capture the standardized FHIR-based core data set modules (resources) defined within the MII working group “Interoperability”. Subsequent steps in the project are the harmonization of the data and biosample sharing governance/regulation frameworks at each ABIDE_MI site, creating synergies for the research infrastructures at the German university hospitals and to link those resources to the German Portal for Medical Research Data and with the BBMRI-ERIC Directory and Negotiator tools.

Burnout syndrome and depression are prevalent mental health problems in many societies today. Most existing methods used in clinical intervention and research are based on inventories. Natural Language Processing (NLP) enables new possibilities to automatically evaluate text in the context of clinical Psychology. In this paper, we show how affective word list ratings can be used to differentiate between texts indicating depression or burnout, and a control group. In particular, we show that depression and burnout show statistically significantly higher arousal than the control group.

National quality measurements with risk-adjusted provider comparison in health care nowadays usually require administrative or clinically measured data. However, both data sources have their limitations. Due to the digitalisation of institutions and the resulting switch to electronic medical records, the question arises as to whether these data can be made usable for risk-adjusted quality comparisons from both a content and a technical point of view. We found that most of the relevant information can be exported with little effort from the electronic medical records. In using this data source an even more sophisticated operationalization of the data of interest is needed.

In this paper we present first findings of the Digi-Care project, a multidisciplinary, multi-stakeholder research project investigating the impacts of digitization on nursing work practices and in particular the transmission of patient care information within and beyond nursing work practices. We completed the initial data collection of the funded 3-year research project and report on a plethora of significant and critical IT-related events. Some of them can be attributed to usability issues.

Digital technology trends for mental health, instantiated with only emerging use cases or already established applications, offer significant potential to improve clinical therapy and care. In this paper, we identify five major trends, mHealth/eHealth, telehealth, artificial intelligence (AI), big data, and biosensors/wearables; describe seven specific technology use cases for mental health care and psychotherapy; and provide an overview of their maturity in practice.

Action observation (AO) and motor imagery (MI) are considered as promising therapeutic approaches in the rehabilitation of patients after a stroke (PaS). Observing and mentally rehearsing motor movements stimulate the motor system in the brain and result in a positive effect on movement execution. To support patients in the early rehabilitation phase after a stroke, ANIMATE, a digital health intervention platform was developed. The platform guides the user through 6 activities of daily living by observing and imagining the corresponding movements. We conducted a scenario-based usability test with 9 PaS at a rehabilitation centre to identify existing usability issues. PaS found the app easy to use and they could interact with it without problems. Although they judged the app as useful, they stated to be not willing to use the app on a regular basis. Including features for customising ANIMATE regarding the individual rehabilitation goals and needs of PaS, as well as personalisation could help in increasing the motivation to use and the benefits of the platform.

Although there are hundreds of mobile yoga apps in the app market space, the quality and usefulness of these apps have not been systematically tested. We conducted a structured quality evaluation of apps from the Google Play store, applying the validated Mobile Application Rating Scale (MARS) by two independent raters. 18 out of 250 apps were identified for evaluation after applying inclusion/exclusion criteria. The mean MARS score is 4.11 (out of 5) with SD = 0.38. There was high interrater reliability (ICC = .88; 95% CI 0.85–0.91). Apps performed well on functionality and aesthetics. However, there is much room for improvement in information and engagement. Designers and researchers should focus on improving user engagement and building the evidence base for informational content provided in apps.

ECG is one of the most common examinations in hospitals, e.g. for diagnosing cardiovascular diseases - the most frequent cause of death worldwide. Goal of this paper was to identify and describe the typical digitized workflow, IT systems and data formats for ECG in hospitals: A survey on current ECG-data management practices was conducted with four German speaking hospitals. A generic model of ECG data management was drafted. Today, these hospitals do not use DICOM as exchange format nor do they implement IHE profiles such as REWF for the ECG. Reasons include missing IT infrastructure such as Master Patient Index or electronic archive. ECG data management could be improved at different levels, with the chance to reduce error sources and to improves in patient safety. Storage of ECG raw data promises better diagnoses based on big data and machine learning technologies.

During resuscitation, the patient is the primary focus with the documentation of actions and outcomes being secondary. In most cases, a cardiac event leads to further treatment or hospitalization, in which complex patient pathways, independent documentation systems and information loss represent the key challenges for successful quality management. Hence, the need for a system that takes all these aspects into account. Market research, system analysis and requirements engineering for such a solution were performed and a prototype was created. A complete reference architecture for a web-based electronic data capture system was developed and implemented that enables healthcare professionals to enter resuscitation-relevant data uniformly and store it centrally in compliance with human research legislation. A qualitative evaluation concerning the process flows of the as-is and the to-be situation suggests that there is potential to achieve benefits in the form of improved data quality and quantity.

Stroke is one of the prevalent diseases which leads to functional disabilities such as hemiparesis or hemiplegia. It is common practice to treat patients with proper rehabilitation as early as possible for better prognosis after the onset of stroke. One of the effective therapeutic techniques for treating stroke patients is mirror therapy, which can potentially facilitate patients’ motor function recovery through repetitive practice. “Rehago” is a software as medical device that implements the concept of mirror therapy in combination with gamified exercises into virtual reality (VR) to provide a home-based rehabilitation environment for stroke patients. In this study, 48 stroke patients completed the full course of intervention with Rehago and their functional performance of pre- and post-intervention was investigated. The intervention with Rehago was predefined as 30 minutes training per day, 5 days per week over a course of 6 weeks. The patient’s progress was evaluated by their therapists every 14 days, with a baseline assessment before the intervention began. The results showed an average improvement of 5.54 points in the Functional Independence Measurement score, and an improvement of 7.13 points in the assessed quality-of-life score (EQ5D-5L). An improvement of the FIM score and the quality-of-life score in EQ5D-5L was observed, indicating it is beneficial to the patients using Rehago as a home-based rehabilitation tool.

Coronary heart disease is among the most frequent causes of death globally. Thus, our research project aims to develop prognostic models, to predict the risk of spontaneous myocardial infarctions based on a combination of clinical parameters and image data sets (invasive coronary angiograms). To train such models we use data from more than 30,000 coronary angiograms acquired at the cardiology department of Erlangen University Hospital. To linking such proprietary data with additional clinical parameters and to harmonize it for future cross-hospital federated machine learning approaches we defined a mapping for coronary angiography based on the symptom/ clinical phenotype HL7^® FHIR^® module of the German medical informatics initiative. In this paper we describe the final design of the coronary angiography information model and our mapping approach to ICD-10 and SNOMED CT. From the database we use a subset of 15 required values patient characteristics to create the HL7^® FHIR^® resource.