Table 2.

Future research theme on development and application AI for digital health and medical analytics

Research themes	Suggested future directions
Sustainable AI	• Policies and practices to leverage AI for socially responsible purposes • Development of curricula and training programs to educate both current and future practitioners • Regulation and legislation to prevent irresponsible application of AI in healthcare services and operations • Agile AI response strategies for future medical crises • Exploring the role of AI in supporting sustainable healthcare systems, e.g., through resource optimisation or predictive analytics for preventive healthcare • Evaluating the long-term sustainability of AI applications in healthcare
Human-centric AI	• Mechanisms to enable user involvement in the design and application of AI solutions • Framework for multi-actor engagement (including AI agents and healthcare professionals) for co-design and evaluation of AI initiatives • Exploration of the paradoxical nature of bias from the socio-materiality view of algorithmic operation • Developing a human-centric approach where healthcare professionals are considered in the loop of AI-enabled clinical decisions • Understanding patient experiences and psychological impacts associated with AI-powered care
Inclusive AI	• Investigation of enablers for inclusive healthcare services using AI • Tactical changes (e.g., role specifications) to employ AI toward inclusive medical services • Integrated communication plans that enhance the experiences of both patients and healthcare professionals • Investigating the role of AI in promoting cultural competence in healthcare, e.g., through language translation or cultural sensitivity algorithms • Assessing potential barriers to AI adoption among diverse populations and developing strategies to overcome them
Fair AI	• Analysis of algorithmic attributes to minimise potential biases in the application of AI, including medical analytics • AI systems that are perceived as trustworthy and fair by key actor groups, primarily patients • Affordable AI solutions to address health disparities among disadvantaged patients • Addressing potential sources of bias in healthcare AI algorithms, such as datasets that lack diversity • Evaluating how AI can contribute to health equity, e.g., by improving access to care or addressing social determinants of health
Transparent AI	• Data management plans to reduce risks of data loss and cyberattacks • Transparent mechanisms for advising evidence-based treatments using AI approaches • AI solutions to make health-related outputs and procedures more accessible to patients