Agents and robots for collaborating and supporting physicians in healthcare scenarios

Monitoring patients through robotics telehealth systems is an interesting scenario where patients' conditions, and their environment, are dynamic and unknown variables. We propose to improve telehealth systems' features to include the ability to serve patients with their needs, operating as human caregivers. The objective is to support the independent living of patients at home without losing the opportunity to monitor their health status. Application scenarios are several, and they spread from simple clinical assisting scenarios to an emergency one. For instance, in the case of a nursing home, the system would support in continuously monitoring the elderly patients. In contrast, in the case of an epidemic diffusion, such as COVID-19 pandemic, the system may help in all the early triage phases, significantly reducing the risk of contagion. However, the system has to let medical assistants perform actions remotely such as changing therapies or interacting with patients that need support. The paper proposes and describes a multi-agent architecture for intelligent medical care. We propose to use the beliefs-desires-intentions agent architecture, part of it is devised to be deployed in a robot. The result is an intelligent system that may allow robots the ability to select the most useful plan for unhandled situations and to communicate the choice to the physician for his validation and permission.