A mental rehabilitation training management system based on doctor-society-family cloud edge collaboration

By using neural control differential equation modeling and an improved sand cat swarm optimization algorithm, the shortcomings of multi-source data fusion and state modeling in existing mental rehabilitation training systems are addressed. This enables the dynamic generation of individualized training plans and risk perception, thereby improving the adaptability and efficiency of training.

CN122201643APending Publication Date: 2026-06-12河南医药大学第二附属医院(河南省精神病医院)
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
河南医药大学第二附属医院(河南省精神病医院)
Filing Date
2026-03-10
Publication Date
2026-06-12

AI Technical Summary

⚠Technical Problem

Existing mental rehabilitation training systems lack multi-source data fusion, have poor continuity in state modeling, and cannot precisely match individual characteristics with training plans. They also cannot achieve dynamic adjustment and process linkage, making it difficult to identify and intervene in sudden abnormal behaviors in a timely manner.

⚗Method used

We employ neural control differential equation modeling, multi-resolution control coding, wavelet decomposition, and an improved sand cat swarm optimization algorithm to construct a cross-scenario, adaptively optimized rehabilitation training generation method. Through data acquisition, preprocessing, wavelet decomposition, control coding, state modeling, model optimization, and training plan generation modules, we achieve the fusion and dynamic collaborative management of multi-source data from medical institutions, communities, and families.

🎯Benefits of technology

It enables continuous modeling and risk perception of multi-source rehabilitation data from medical institutions, communities, and families, supports the dynamic generation of individualized rehabilitation training plans, improves the adaptability and execution efficiency of training, and enhances the model's comprehensive performance in adapting to multi-objective rehabilitation needs, state prediction, compliance fitting, and risk identification.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122201643A_ABST
    Figure CN122201643A_ABST
Patent Text Reader

Abstract

The application discloses a mental rehabilitation training management system based on doctor-society-family cloud edge cooperation, which comprises a data acquisition and preprocessing module, a wavelet decomposition processing module, a control coding module, a state evolution modeling module, a model optimization module, a training plan generation module and a task issuing module. The data acquisition and preprocessing module generates a doctor-society-family control sequence. The wavelet decomposition processing module extracts a multi-resolution sub-sequence. The control coding module outputs a doctor-society-family control embedding. The state evolution modeling module generates a mental rehabilitation hidden state sequence. The model optimization module optimizes structural parameters. The training plan generation module reasons and corrects the hidden state sequence and formulates an individualized rehabilitation training plan. The task issuing module sends the training plan to a doctor-society-family edge node through a cloud edge cooperation topology gate layer. The mental rehabilitation training intelligent optimization is realized, and the model adaptability and intervention response efficiency are improved.
Need to check novelty before this filing date? Find Prior Art