Follow-up visit monitoring method and device based on federal reinforcement learning and related equipment

Abstract

The invention relates to an artificial intelligence technology, and provides a follow-up visit monitoring method and device based on federal reinforcement learning, computer equipment and a storage medium, and the method comprises the steps of obtaining an initial follow-up visit monitoring model and a cluster in a remote server; training the initial follow-up monitoring model according to the follow-up data to obtain a first follow-up monitoring model; encrypting a first model parameter of the first follow-up monitoring model to obtain an encrypted first model parameter set; outputting the encrypted first model parameter set to a far-end server for weighting to obtain a second model parameter; receiving a second model parameter output by the remote server, and weighting the first model parameter and the second model parameter to obtain a target model parameter; updating the first follow-up monitoring model according to the target model parameter to obtain a target follow-up monitoring model; and calling the target follow-up visit monitoring model to process follow-up visit data of the target user to obtain a follow-up visit monitoring path. Accuracy of follow-up visit monitoring can be improved, and rapid development of a smart city is promoted.

Description

technical field [0001] The present application relates to the technical field of artificial intelligence, and in particular to a follow-up monitoring method, device, computer equipment and media based on federated reinforcement learning. Background technique [0002] Recently, the application of artificial intelligence technology in chronic disease management has made great progress. In the chronic disease management scenario, the biggest contradiction is the balance between effective medical resources and huge patient follow-up needs. For now, the optimal strategy optimization solution is to build a reinforcement learning model through the massive data collected, and automatically allocate medical resources through the reinforcement learning model for follow-up patients who need intervention from medical resources. [0003] In the process of realizing the present application, the inventor found that the following technical problems exist in the prior art: patients with dif...

AI Technical Summary

Problems solved by technology

[0003] In the process of realizing the present application, the inventor found that the following technical problems exist in the prior art: patients with different conditions such as different age groups and different degrees of illness will have different tendencies in the follow-up intervention program, for example, the elderly are more It is hoped that medical staff will intervene, rather than intelligent unmanned intervention on the mobile phone

In the follow-up scenario, many patients often do not upload their personal data due to safety considerations, resulting in a serious lack of follow-up data, which makes the training of the reinforcement learning model lack data support, and cannot guarantee the accuracy of follow-up monitoring

Images