Enterprise AI Analysis
Personalised Health Plan Development Using Agentic AI in Singapore's National Preventive Care Programme: A Pilot Study
This pilot study demonstrates the feasibility and positive user acceptance of an agentic AI digital assistant, HealthGuide@Home, for generating personalised health plans within Singapore's national preventive care programme. By leveraging multi-agent frameworks, the system refines health plans based on user interactions and preferences, addressing the critical need for scalable, individualised health guidance amid workforce shortages.
Executive Impact: Key Metrics & ROI
Our analysis highlights significant potential for enhanced user engagement and health outcomes through agentic AI. The pilot revealed strong positive sentiment and high acceptance across key metrics, indicating readiness for broader enterprise integration.
0
Positive User Acceptance
0
Personalisation Valued
0
Granularity Appreciation
0
Positive Feedback Sentiment
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
AI Foundation & System Design
The HealthGuide@Home system utilizes a state-of-the-art agentic framework, LangGraph, selected for its scalability, modularity, and memory management, crucial for dynamic task decomposition and continuous adaptation in health plan generation. Claude 3.5 Sonnet was chosen as the core LLM engine due to its superior cognitive performance, function calling, instruction adherence, and flexibility, vital for complex healthcare tasks. The system's semantic router workflow orchestrates preference gathering, personalization, option discovery, and feedback collection to iteratively refine health plans based on user input and trusted local data sources, including official MOH/HPB guides and open-source knowledge bases.
Positive User Reception & Validation
Both residents and clinicians rated the AI-generated health plans significantly above neutral satisfaction across Appropriateness, Usefulness, Actionability, and Personalisation (all p-values < 0.05). Residents highly valued personalisation (p=0.003) and the granularity of recommendations (p=0.0003), with no major concerns about AI-generated plans. Sentiment analysis revealed predominantly positive feedback for personalized diet, exercise, and general features, especially after plan refinement, highlighting the system's ability to align with individual needs and preferences.
Strategic Considerations & Roadmap
While promising, the pilot identified key areas for future development. Limitations include small sample size and lack of direct feedback from patients with complex health conditions. Addressing AI trustworthiness requires robust frameworks for safety, privacy, bias mitigation, and interpretability. Future directions involve enhancing user engagement through gamification and visual aids, integrating a comprehensive risk stratification framework for patients with comorbidities, and conducting long-term effectiveness studies to validate sustained behaviour change and health outcomes within national initiatives like Healthier SG.
0
of Residents Valued Personalisation in Health Plans (p-value: 0.003)
Enterprise Process Flow: HealthGuide@Home Semantic Router
Preference Gathering
→
Personalisation
→
Option Discovery
→
Feedback Collection
| Capability | Claude 3.5 Sonnet | Llama 3 70B Instruct |
|---|---|---|
| Cognitive Performance | ✓ Superior (4.9 vs 4.7) | |
| Function Calling | ✓ Superior (5.0 vs 4.6) | |
| Instruction Adherence | ✓ Superior (4.9 vs 4.1) | |
| Flexibility | ✓ Superior (4.0 vs 3.5) | |
| Faithfulness (Factual Accuracy) | ✓ Higher (0.903 vs 0.887) | |
| Answer Relevancy (User Alignment) | ✓ Slightly Better (0.858 vs 0.848) | |
| Fine-Tuning for Domain Specificity | ✓ Superior | |
| Processing Speed (Real-time Use) | ✓ Faster |
Real-world Pilot: Agentic AI in Singapore's Preventive Care
Problem: Singapore, like many nations, faces significant demographic shifts leading to an aging population and increasing demand for healthcare services, especially preventive care for chronic diseases. Existing workforce shortages limit the ability to provide extensive, personalized health plans at scale.
Solution: HealthGuide@Home, a digital assistant powered by a multi-agent AI framework, was implemented to generate and refine personalized health plans based on user interactions, preferences, and clinical insights. This system aims to support the Healthier SG national initiative by shifting from reactive to proactive health management.
Outcome: A pilot study with 20 residents and 7 clinicians demonstrated positive user acceptance. Key findings include high value placed on personalization and granularity, with positive sentiment across diet, exercise, and general health plan features. This highlights agentic AI's potential to augment preventive care, empower patients, and optimize healthcare resource utilization by delivering scalable, personalized guidance.
Quantify Your AI Impact
Estimate the potential savings and reclaimed hours for your enterprise by implementing similar AI-driven solutions for personalized health management.
Estimated Annual Savings
$0
Annual Hours Reclaimed
0
AI Implementation Roadmap
A phased approach to integrate agentic AI for personalized health plans within your organization, ensuring strategic alignment and measurable outcomes.
Phase 1: Strategy & Pilot Program
Align AI solution with existing clinical guidelines and organizational health objectives. Conduct a controlled pilot with a focused user group to gather initial feedback and validate system effectiveness and user acceptance in a real-world setting.
Phase 2: Platform Integration & Refinement
Seamlessly integrate the AI digital assistant with existing healthcare IT infrastructure and patient management systems. Implement iterative feedback loops for continuous refinement of AI models, enhancing personalization, accuracy, and user experience based on pilot insights.
Phase 3: Scalable Deployment & Monitoring
Roll out the personalized health plan solution to a broader population, ensuring robust risk stratification and safety guardrails. Establish continuous monitoring for health outcomes, user engagement, and AI performance, with ongoing validation against clinical standards and adaptation to evolving needs.
Ready to Transform Healthcare with AI?
Schedule a consultation with our experts to explore how agentic AI can revolutionize your preventive care programs and drive measurable health outcomes.
Ready to Get Started?
Book Your Free Consultation.
Let's Discuss Your AI Strategy!
Lets Discuss Your Needs
Select a Date
Sun
Mon
Tue
Wed
Thu
Fri
Sat