Self-adaptive infusion cooperative control method and system based on intelligent decision

By combining an end-to-end patient status assessment model and an LSTM adaptive flow rate control algorithm with federated learning, the problem of flow rate control accuracy and stability of existing infusion devices in complex environments has been solved. This has enabled personalized infusion control and collective intelligent sharing, improving the intelligence level and data privacy protection capabilities of the medical system.

CN122097753AActive Publication Date: 2026-05-29WEIFANG UNIVERSITY
12 Cites 0 Cited by

Patent Information

Application Number
CN202610568781.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-04-28
Publication Date
2026-05-29
Estimated Expiration
2046-04-28

AI Technical Summary

Technical Problem

Existing infusion equipment is difficult to implement in environments lacking suspension fulcrums, has low flow rate control accuracy and poor stability, cannot deeply integrate with the patient's physiological state to make personalized treatment decisions, and lacks a collaborative mechanism between devices, making it difficult to improve the overall diagnostic and treatment efficiency and intelligence level, and the problem of medical data privacy protection remains unresolved.

Method used

An adaptive infusion collaborative control method based on intelligent decision-making is adopted. By integrating an end-to-end patient status assessment model and an LSTM-based adaptive flow rate control algorithm, combined with a federated learning framework, model parameter sharing and privacy protection are achieved among multiple devices. A three-level collaborative network of "cloud-edge-device" is constructed to carry out personalized and safe infusion control.

Benefits of technology

It achieves precise control over individual patients and knowledge sharing among groups, improving the safety and intelligence of infusion. It can automatically maintain the accuracy and safety of treatment under complex conditions, is suitable for large-scale emergency scenarios, and has global situational analysis capabilities to respond to public health emergencies.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to the technical field of medical internet of things, in particular to a self-adaptive infusion cooperative control method and system based on intelligent decision making. The method comprises the following steps that: end equipment analyzes multi-modal physiological time sequence data and infusion parameters through an end-to-end patient state evaluation model integrated on a sub-body infusion unit, and outputs a patient state index and an infusion suggestion; and in combination with a pipeline pressure signal, a motor control instruction is generated through an LSTM-based adaptive flow speed control algorithm, and closed-loop control is realized. The side device periodically aggregates local model parameter updates of the plurality of end devices through the parent base station. And the cloud device fuses updates uploaded by all the edge devices based on a federated average algorithm, generates an optimized global model and issues the optimized global model. According to the method, normal form transformation of infusion from isolated execution to collaborative awareness-decision-evolution is realized, intelligent continuous evolution of a group is driven on the premise of protecting data privacy, and the accuracy, the safety and the overall intelligent level of infusion treatment in a complex environment are improved.
Need to check novelty before this filing date? Find Prior Art