Curious thing, an artificial intelligence (AI)-based conversational agent for COVID-19 patient management.

Area of Science:

• Digital Health
• Artificial Intelligence in Healthcare
• Public Health Surveillance

Background:

• Lack of validated models for AI in monitoring community-isolated COVID-19 patients.
• Need for efficient remote monitoring solutions for low-to-medium risk patients.

Purpose of the Study:

• To evaluate an AI conversational technology (Curious Thing) for monitoring home-isolated COVID-19 patients.
• To assess the effectiveness of AI in identifying patients requiring further support and optimizing clinical resources.

Main Methods:

• Deployment of an AI care robot (Curious Thing) in South Western Sydney from September to December 2021.
• AI-assisted phone calls using a local script to gather patient health, wellness, and support information.
• Monitoring of over 6,323 low-to-medium risk COVID-19 patients isolating at home.

Main Results:

• AI-assisted calls successfully identified patients needing additional support.
• Clinician time was saved by remotely monitoring less acute patients.
• Resources were effectively directed to critically ill patients, improving care allocation.

Conclusions:

• AI-driven conversational technology provides an effective model for remote patient monitoring during home isolation.
• The technology enhanced clinical efficiency and patient-centered virtual care delivery.
• Positive feedback from patients and staff supports the ongoing development of virtual care models.