DECIDE-Twin: A Framework for AI-Enabled Digital Twins in Clinical Decision-Making

Area of Science:

• * Digital Health
• * Artificial Intelligence
• * Clinical Decision Support

Background:

• * AI-enabled digital twins (DTs) offer transformative potential in healthcare decision-making.
• * A unified framework for DT development and implementation is currently lacking in the literature.
• * Growing interest necessitates structured approaches for AI-driven virtual models in clinical settings.

Purpose of the Study:

• * To systematically map existing knowledge on AI-enabled DTs for clinical decision-making.
• * To develop a comprehensive framework for the creation and deployment of these advanced virtual models.
• * To address the gap in unified guidance for AI-DTs in healthcare.

Main Methods:

• * Conducted a scoping review following established frameworks (Arksey and O'Malley, JBI).
• * Searched multiple databases (PubMed, Embase, IEEE Xplore) and grey literature until August 2023.
• * Employed dual reviewer screening and content analysis to synthesize findings and develop the DECIDE-Twin framework.

Main Results:

• * Included 11 studies (7 reviews, 4 empirical) published from 2020 onwards.
• * Identified key development steps from existing frameworks and empirical studies.
• * Synthesized information to create the novel DECIDE-Twin framework.

Conclusions:

• * The DECIDE-Twin framework provides a structured guide for developing and implementing AI-enabled DTs in clinical decision-making.
• * This framework can assist both researchers and practitioners in the healthcare domain.
• * Further validation and application across diverse clinical use cases are recommended.