Project

Back to overview

Ethical and Legal issues of Mobile Health-Data - Improving understanding and eXPlainability of digitaL transformAtion and data technologies using artificial IntelligeNce [EXPLaiN]

English title Ethical and Legal issues of Mobile Health-Data - Improving understanding and eXPlainability of digitaL transformAtion and data technologies using artificial IntelligeNce [EXPLaiN]
Applicant Elger Bernice
Number 187263
Funding scheme NRP 77 Digital Transformation
Research institution Institut für Bio- und Medizinethik (IBMB) Universität Basel
Institution of higher education University of Basel - BS
Main discipline Health Education
Start/End 01.10.2020 - 30.09.2024
Approved amount 458'100.00
Show all

All Disciplines (6)

Discipline
Health Education
Health
Sociology
Philosophy
Legal sciences
Public Health and Health Services

Keywords (5)

Research ethics; Machine learning; Mobile health data; Stakeholder attitudes; Legal framework

Lay Summary (German)

Lead
EXPLaiN untersucht die ethischen und rechtlichen Fragen, die der Umgang mit mobile Gesundheutsdaten aufwirft. Ziel ist, das Verständnis und die Erklärbarkeit der digitalen Technologien, insbesondere der Verwendung von Künstlicher Intelligenz (KI) bei der Analyse von Gesundheitsdaten, zu verbessern.
Lay summary

Die Themen mobile Gesundheitsdaten (mDaten) und Künstliche Intelligenz (KI) sind eng miteinander verbunden. Das Volumen von mDaten ist explosiv in die Höhe geschnellt, seit Apps ermöglichen EGK, Puls, Bewegungs- und viele andere Gesundheitsrelevante Daten mithilfe von sogenannten Wearables zu erheben. Methoden der KI wie Machine leraning (ML) sind vielverspechend für die Analyse dieser Datenvolumina. EXPLaiN kombiniert verschiedene Methoden, um ethische, rechtliche und technologiebedingte Fragen bei der Erhebung und Nutzung von mDaten zu klären. Am Beispiel von Daten in Anästhesie und Kardiologie wird die Entwicklung von ML Algorithmen medizinisch, ethisch und rechtlich begleitet, um Datenschutz, Erklärbarkeit der Analysen, und Bedenken von medizinischem Personal und von Patient*innen besser zu verstehen.

Direct link to Lay Summary Last update: 31.01.2020

Responsible applicant and co-applicants

Employees

Collaboration

Group / person Country
Types of collaboration
Dr. Philipp Egli Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
Prof. Christian Sticherling Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
Prof. Leo Anthony Gutierrez Celi, M.D. United States of America (North America)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
Prof. Marie-Laure Papaux Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
Prof. Thomas Perneger Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
Prof. Christian Kern Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
Maître Jean-François Dumoulin Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
Prof. Ezekiel Emanuel United States of America (North America)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
Prof. Henning Müller Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
Jean-Daniel Zeller Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
Bruno Baeriswyl Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
Dr. (PhD) Jörg Willers Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
Prof. Frank Stüber Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
Prof. Dipen Shah Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
Prof. Thomas Gächter Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication

Abstract

Research topic and background: This project explores the highly important topic of mobile data (mData) uses in medicine and the employment of machine learning (ML) and deep neural networks (DNN) in this context. Clinical, research or other secondary uses involving mobile health data or ML, and even more the combination of mData and ML, have raised a panoply of new concerns creating legal and ethical barriers which interfere with trust of patients and society. Data protection concerns increase disproportionally with mData. In addition, the use of ML could be incompatible with the new EU General Data Protection Regulation (GDPR, Art. 22) that postulates a right to explanation while ML algorithms function like a “black box” in spite of efforts to improve interpretability and explainability. Given the complexity of regulatory issues related to mData and ML, research ethics committees (REC) worldwide struggle with establishing criteria how to approve such projects.These concerns significantly impede advances in research: health care institutions collect high volumes of very useful data, including an increasing amount of mData, but beneficial analysis, in particular of mobile data, remains scarce. While appropriate infrastructures are currently being developed, there is an urgent need for (i) clarification of the new pressing questions related to the ethical and legal governance of mobile data research and the use of ML algorithms and (ii) an in depth exploration of patient concerns.The topics of mData and ML are intrinsically related as the volume of mData collected during health care has recently started to explode. Examples are smartwatch apps used in cardiology and non-invasive home-ventilators (continuous positive airway pressure, CPAP) collecting a variety of sleep, breathing and activity related parameters. Artificial intelligence (AI) such as ML is a promising tool to analyze this high data volume for research or clinical purposes.While the global importance of AI technology is generally recognized for the interpretation of digitalized data in the clinical and/or research context - f. ex. scans, skin lesions, electrocardiograms, faces, and vital signs - challenges such as bias, data protection, and lack of transparency and explainability raise concern. Access to data is debated, as manufacturers and health insurers are highly interested in the use of mData. These unresolved issues interfere with trust and the efficient and beneficial implementation of these promising technologies.Ethical and legal issues, including risks and benefits related to ML in medicine, vary enormously between different examples. Therefore, it is important to study them in concrete applications. This project will fill an important gap through a pragmatic approach on mData research and ML, by examining these questions related to two existing types of data collected in Swiss University hospitals. The first type concerns mobile data collected in cardiac patients via smartwatches. The second type consists of perioperative data (POD) collected via existing patient data management systems: PDMS, Copra 6 at Swiss University hospitals (BS, BE; VD in prep.). Both datasets are producing high amounts of interoperable data (ECG, vital signs etc.) where ML is highly useful for research and clinical purposes. The two datasets will be compared to examine which type of additional issues exist for the use of mData as compared to clinical and secondary uses for more classical in-hospital data.Objectives and milestones:i. Define needs and barriers in practice to advance data analysis in two case studies- Use two existing types of data collected in Swiss hospitals (mData from smart watches used in cardiology at the University Hospital Basel and POD data from PDMS) to (a) describe scientifically meaningful clinical and research or other secondary uses and (b) clarify the ethico-legal framework and ML explainability to advance cutting-edge analysis. Our applicants from the field of computer science will contribute their experience on explainability as they not only use ML algorithms to build predictive models from biomedical data, but also investigate approaches allowing to understand reasons behind the models.ii. Review the existing ethical and legal frameworks- Analyse the current research ethics framework for data analysis involving ML, in particular for new types of mobile data collected via apps on smart watches.- Analyse the legal framework, in Switzerland and Europe (in particular the European data protection regulation: GDPR), including (a) existing law and how it should be interpreted, (b) legal barriers or gaps and (c) a de lege ferenda approach, i.e. how the law and research ethics framework should regulate mobile data and the use of AI/ML.iii. Obtain new Swiss data on trust-enhancing factors and solutions through qualitative interviews with experts and patients to understand attitudes and information needs of the persons whose data will be used, and to explore opportunities and barriers to consent to data research and ML.iv. Combine results from all parts i.-iii. to clarify regulatory and patient trust-related concerns about the use of their data and ML for the two databases and disseminate results through educational tools.- Use ethical analysis, to influence the local and international regulatory debate, to further a sensitive and efficient research ethics framework and to facilitate analysis using AI/ML of other mData and hospital databases.- Propose, after thorough review of the relevant ethico-legal literature, a framework consistent with present legal requirements and/or make recommendations for the harmonisation or adaptation of regulations- Disseminate the findings and recommendations through publications, conferences and teaching.
-