Research On Intelligent Rehabilitation Planning Method For The Disabled

With the acceleration of population aging process in China,the health care of the elderly has attracted more and more attention.The prevalence of chronic noncommunicable diseases represented by disability,mental illness,diabetes,cardiovascular and cerebrovascular diseases,cancer and chronic respiratory diseases has been increasing year by year.Among them,disability has become a thorny issue for the elderly in the field of disability rehabilitation due to its high incidence and great harm.At present,There are many problems for the rehabilitation planning research in the field of disability,such as the lack of real data,the interference of dirty data,and the lack of standard samples.The lack of data sets restricts the research of related intelligent algorithms.In addition,traditional disability rehabilitation services rely on a large number of offline rehabilitation institutions,which also leads to problems such as uneven resource allocation,low service efficiency,and low intelligence.In response to the above problems,this paper builds three basic elements: disability rehabilitation data sets,data-driven intelligent rehabilitation planning and knowledge graph-based intelligent rehabilitation planning,and the relevant theoretical algorithms are studied.The content is as follows:First of all,in response to the problem of small sample learning in the field of disability rehabilitation,this paper proposes a three-stage disability rehabilitation data generation model based on machine learning.The first stage designs a basic attribute generation strategy based on a tree structure;the second stage builds a native Bayesian basic behavior ability indicator generation model;the third stage design uses multiple linear regression model to generate high-order behavior ability indicators.In the end,the three phases of work are integrated,and the generation and expansion of the disability rehabilitation data set is realized.Secondly,aiming at the low efficiency and high cost of traditional rehabilitation planning methods,this paper proposes a data-driven intelligent rehabilitation planning method.At first this paper designs and implements a neural network-based feature extraction model for the feature extraction of samples.Then,the rehabilitation planning is split according to internal data items and divided into different classification problems,and a model based on the classification network is constructed.After getting feature vector,it is used as the input of the model to obtain the classification result,which is combined and recommended to the user.Third,aiming at the problems of poor knowledge structure and lack of knowledge in the field of disability rehabilitation,this paper proposes an intelligent rehabilitation planning method based on knowledge graphs.At first this article determines the topdown graph construction logic,combines expert knowledge to define the ontology of the disability rehabilitation field,and uses a Web page wrapper-based approach to extraction information.After completing the knowledge fusion,it is stored in the native graph database Neo4 j.Based on the construction of the graph,the application of knowledge search is designed and realized.Finally,on the basis of the above theoretical methods,this paper designs and implements the intelligent rehabilitation planning subsystem,and verifies the feasibility and effectiveness of the research content through systematic analysis and testing.