Smart health-care is the most promising application of the next-generation 5G wireless network. Because of low latency and high data rate, many applications with high resources are supporting 5G, including smart health-care application. In smart health-care, medical sensors exchange data to establish a network. However, the mobility of nodes and density changes the network topology usually. Medical sensor nodes have limited energy, which is used for transmission and receiving of data. In this paper, an idea of selection of route is distinguished by taking into account of stability and higher residual energy in 5G-based smart health-care network to decrease energy consumption along with links disconnection and improve network lifetime. For this purpose, we present reinforcement learning-based algorithm and investigate the effect of various learning rates on energy consumption, links disconnection and network lifetime in smart health-care network.