Personalised health plan development using agentic AI in Singapore's national preventive care programme: a pilot...

Area of Science:

• Digital Health
• Artificial Intelligence
• Preventive Medicine

Background:

• Aging populations create workforce shortages in healthcare.
• A shift towards preventive healthcare strategies is necessary.
• Generative AI offers potential for personalized health guidance.

Purpose of the Study:

• To implement and evaluate a digital assistant using a multi-agent framework for personalized health plans.
• To assess user acceptance and satisfaction with the AI-driven system.

Main Methods:

• Developed a multi-agent framework digital assistant for generating and refining personalized health plans.
• Conducted a pilot study with 20 residents and 7 clinicians.
• Collected feedback on user acceptance and satisfaction with key features.

Main Results:

• Both residents and clinicians reported satisfaction above neutral levels (p < 0.05).
• Residents valued personalization (p=0.003), granularity (p=0.0003), and had few concerns (p=0.941).
• Positive sentiment was observed for personalized diet (p=0.110), exercise (p=0.003), and general features (p=6e-06).

Conclusions:

• AI-driven digital assistants show potential in supporting preventive healthcare programs.
• The multi-agent framework is a viable approach for personalized health plan generation.
• User acceptance indicates the feasibility of integrating such technologies into healthcare.