From Biometrics to Environmental Control: AI-Enhanced Digital Twins for Personalized Health Interventions in Healing Landscapes

Real-time health monitoring enables personalized environmental adjustments: The system uses ECG data and heart rate variability measurements to continuously detect stress levels in individuals, then automatically adjusts temperature, humidity, and ventilation to match their physiological needs—moving beyond one-size-fits-all environmental controls.

Explainability is built into the decision-making process: The researchers used SHAP (Shapley Additive exPlanations) to interpret how the AI model makes stress predictions, identifying which specific biometric features drive each decision. This transparency helps users and healthcare providers understand why particular environmental interventions are recommended.

Multi-level interventions scale from personal to landscape: Rather than only adjusting a single room's temperature, the framework can activate coordinated responses across personal devices, individual rooms, entire buildings, and outdoor landscapes—allowing health interventions to work at multiple scales simultaneously.

The system was validated on established medical datasets: Testing used the MIT-BIH noise stress test dataset, a recognized benchmark in cardiac research, which provides credibility that the stress detection approach is grounded in real physiological data rather than theoretical assumptions.

Privacy and data governance considerations are implicit in the design: The integration of continuous biometric monitoring with environmental control systems raises important questions about data collection, storage, and consent that responsible deployment would need to address through clear policies and user control mechanisms.