Resource Allocation Management in Patient-to-Physician Communications Based on Deep Reinforcement Learning in Smart Healthcare Services

The integration of smart cities and healthcare has led to the use of technology and information in medical practices around the world. In this study, we improve the mechanism of resource allocation for communications in the form of a decentralized patient-to-physician (P2P) scheme. We studied the capabilities of deep reinforcement learning to enhance the resource allocation model to optimally exploit resources, and thus fulfill the quality of service. We developed a decentralized resource allocation management model in P2P communications according to deep reinforcement learning based on multi-agents, wherein the constraints of the delay on P2P links can be addressed directly. In accordance with the effects of recreation, every operator can proficiently identify how to satisfy the severe limitations of dormancy on P2P link correspondence, and concurrently decrease the obstructions to the patient/doctor to-infrastructure link interchanges.