Ethical use of artificial intelligence to prevent sudden cardiac death: an interview study of patient perspectives.

Autor: Maris MT; Department of Ethics, Law and Humanities, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands. m.t.maris@amsterdamumc.nl., Koçar A; Institute for Healthcare Management and Health Sciences, University of Bayreuth, Bayreuth, Germany., Willems DL; Department of Ethics, Law and Humanities, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands., Pols J; Department of Ethics, Law and Humanities, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands.; Department of Anthropology, University of Amsterdam, Amsterdam, The Netherlands., Tan HL; Department of Clinical and Experimental Cardiology, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands.; Netherlands Heart Institute, Utrecht, The Netherlands., Lindinger GL; Institute for Healthcare Management and Health Sciences, University of Bayreuth, Bayreuth, Germany., Bak MAR; Department of Ethics, Law and Humanities, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands.; Institute of History and Ethics in Medicine, TUM School of Medicine, Technical University of Munich, Munich, Germany.
Jazyk: angličtina
Zdroj: BMC medical ethics [BMC Med Ethics] 2024 Apr 04; Vol. 25 (1), pp. 42. Date of Electronic Publication: 2024 Apr 04.
DOI: 10.1186/s12910-024-01042-y
Abstrakt: Background: The emergence of artificial intelligence (AI) in medicine has prompted the development of numerous ethical guidelines, while the involvement of patients in the creation of these documents lags behind. As part of the European PROFID project we explore patient perspectives on the ethical implications of AI in care for patients at increased risk of sudden cardiac death (SCD).
Aim: Explore perspectives of patients on the ethical use of AI, particularly in clinical decision-making regarding the implantation of an implantable cardioverter-defibrillator (ICD).
Methods: Semi-structured, future scenario-based interviews were conducted among patients who had either an ICD and/or a heart condition with increased risk of SCD in Germany (n = 9) and the Netherlands (n = 15). We used the principles of the European Commission's Ethics Guidelines for Trustworthy AI to structure the interviews.
Results: Six themes arose from the interviews: the ability of AI to rectify human doctors' limitations; the objectivity of data; whether AI can serve as second opinion; AI explainability and patient trust; the importance of the 'human touch'; and the personalization of care. Overall, our results reveal a strong desire among patients for more personalized and patient-centered care in the context of ICD implantation. Participants in our study express significant concerns about the further loss of the 'human touch' in healthcare when AI is introduced in clinical settings. They believe that this aspect of care is currently inadequately recognized in clinical practice. Participants attribute to doctors the responsibility of evaluating AI recommendations for clinical relevance and aligning them with patients' individual contexts and values, in consultation with the patient.
Conclusion: The 'human touch' patients exclusively ascribe to human medical practitioners extends beyond sympathy and kindness, and has clinical relevance in medical decision-making. Because this cannot be replaced by AI, we suggest that normative research into the 'right to a human doctor' is needed. Furthermore, policies on patient-centered AI integration in clinical practice should encompass the ethics of everyday practice rather than only principle-based ethics. We suggest that an empirical ethics approach grounded in ethnographic research is exceptionally well-suited to pave the way forward.
(© 2024. The Author(s).)
Databáze: MEDLINE
Externí odkaz: