Perspectives on the use of artificial intelligence in Japan: a focus group interview study of healthcare providers

Abstract

Introduction: The integration of artificial intelligence (AI) into healthcare is accelerating, raising important questions regarding its implications for clinical practice and the roles of healthcare providers (HCPs). Despite significant technical advances, strategies for engaging frontline stakeholders in developing and implementing medical AI remain underexplored. Therefore, this qualitative study aimed to understand the views and perspectives of stakeholders regarding the use of AI in healthcare. Methods: We examined the perspectives of 37 healthcare professionals (doctors, nurses, and allied HCPs) in Japan through a series of focus group interviews conducted in 2022. Participants discussed three clinical scenarios involving AI technologies for lung cancer detection, voice recognition during consultations, and healthcare monitoring. A thematic analysis was conducted to explore the views of HCPs on the use of AI in healthcare. Results: Diverse and sometimes contrasting views were obtained on the benefits, risks, and practical challenges associated with AI in healthcare. Participants reflected on issues such as algorithmic accuracy, bias, responsibility, and the risk of over-reliance. They also raised fundamental questions regarding how AI might redefine the concepts of disease, patient autonomy, and the boundaries between healthcare and everyday life. Although AI was generally considered a support tool for clinical judgement, concerns were raised regarding its potential to reshape healthcare workflows, exacerbate inequalities, and shift professional roles. Conclusion: The results highlight the importance of involving stakeholders in the early stages of AI development to anticipate its broader implications. Beyond regulating AI as a device, designing healthcare systems that are adaptable, inclusive, and human-centred is necessary. This study provides empirical insight into how HCPs conceptualise AI and emphasises the need for anticipatory governance frameworks grounded in real-world practice.

Ähnliche Arbeiten