AI-Guided Computing Insights into a Thermostat Monitoring Neonatal Intensive Care Unit (NICU)

Area of Science:

• Healthcare Engineering
• Computational Science
• Artificial Intelligence

Background:

• Effective airflow and ventilation are critical in healthcare settings, particularly neonatal intensive care units (NICUs).
• Monitoring and controlling environmental parameters like temperature are essential for infant well-being.

Purpose of the Study:

• To investigate airflow and heat transfer within a neonatal intensive care unit (NICU) using computational fluid dynamics (CFD).
• To develop and implement an artificial intelligence (AI) model, incorporating machine learning (ML), for real-time prediction of thermal performance in NICU designs.

Main Methods:

• Modeled a single-patient NICU room based on realistic dimensions and standards.
• Employed Navier-Stokes equations to simulate incompressible fluid flow and heat transfer.
• Integrated an AI/ML model using CFD geometric parameters and thermophysical variables (mass flow rate, heat flux, temperature gradient) for predictive analysis.

Main Results:

• CFD simulations revealed expected airflow patterns and heat transfer dynamics.
• The AI model accurately predicted thermal performance in real-time.
• Inclusion of thermophysical variables enhanced the simulation's accuracy in capturing convective heat transfer and temperature gradients.

Conclusions:

• AI and ML models can effectively process large datasets generated in NICU environments.
• These AI-driven approaches can identify patterns and assist with sensor data interpretation for improved patient care.
• Physics-informed AI enhances the generalizability and numerical accuracy of fluid flow and heat transfer simulations.