| In recent years,due to the economic level and social structure having different degrees of impact on human health,the disease modes are complex and diverse,and the health problems of young and middle-aged people have become increasingly prominent.The world health data shows that there is a clear trend of age decline for various chronic diseases such as hypertension and hyperlipidemia.In recent years,the prevalence of chronic diseases among young and middle-aged people in my country has increased exponentially.On the one hand,young and middle-aged people shoulder the mission of national construction;on the other hand,young and middle-aged people bear the responsibility of providing for the elderly and raising offspring.The health of young and middle-aged people affects the overall national quality.With the advancement of national intelligence reforms and the improvement of the quality of life,people pay more attention to the early warning of diseases.Machine learning methods play an important role in the operation of early warning models and are now widely used in multidisciplinary fields such as management and medicine.The young and middle-aged people are the key population to improve the overall health level of our country,and there is still a lack of research on combining the health status of the young and middle-aged residents in our country with machine learning methods for disease warning,and it is urgent to research and tap the value.Based on the concept of improving the overall health of young and middle-aged people in my country,this research uses an intelligent health monitoring system to collect data,analyzes the health status of young and middle-aged residents,and builds an optimized medium based on BP neural network and SVM(Support-vector Machine).The disease early warning model for young residents provides a scientific basis for formulating relevant health intervention measures.The specific work is divided into four parts:First,based on domestic and foreign historical documents and policy requirements,summarize the background,significance and urgency of the research,clarify the research objects,and propose research questions and research plans.Asked many experts in the field for suggestions on the design of research plans,combined with previous literature research and the system design of the research team,to construct indicators and evaluation models for evaluating the health of young and middle-aged people.The literature reviews the current status of chronic diseases of young and middle-aged people at home and abroad,the assessment content of health status,and the application of early warning models in medicine.Second,using the intelligent health monitoring system as a questionnaire distribution tool,it is aimed at Chinese middle-aged and young residents aged 18 to 60,using convenient sampling to collect 4178 health monitoring data,reviewing the questionnaire and establishing a database.Using descriptive statistics to analyze the general characteristics of young and middle-aged residents,behaviors affecting their health,and disease status,it was found that the average age of this study was(27.99±9.631)years old,and the age group was mainly distributed between 18 and 32 years old.The main body of this survey;the smoking rate,passive smoking rate,and alcohol drinking rate were 11.3%,47.8%,and 32.0%,respectively;the chronic disease rate was 12.4%.Chronic gastritis was the chronic disease with the highest prevalence rate(4.7%),and there were anxiety and The proportions of depressive symptoms were 35.0%and38.3%,respectively.Third,using theχ~2 test and the rank sum test were used to do a single factor analysis of general characteristics,behaviors affecting health status,and disease status,and 40 statistically significant indicators were determined as input variables of the disease warning model for young and middle-aged residents;Matlab R2019b software was used,Preliminarily build a BP neural network model to learn and predict the health data of young and middle-aged residents.The overall accuracy of the initial model to predict the disease status of young and middle-aged residents is 94.6%;Because the BP neural network has certain limitations,a disease warning model based on SVM is constructed and the prediction results are integrated,and the comprehensive accuracy rate is 98.56%.Fourth,in addition 10 samples were selected to test the final disease warning model for young and middle-aged residents,and the test results were displayed and analyzed.There were9 samples with correct predictions and 1 sample with incorrect predictions.Based on the analysis of the results of this study,corresponding recommendations should be put forward.This study found that the adjustment of chronic gastritis,passive smoking,mild anxiety and depression in the young and middle-aged population needs more attention.The final model can improve the accuracy and effectiveness of early warning of diseases of young and middle-aged people.Normalized health monitoring should be strengthened for early warning of chronic diseases.In daily life,attention should be paid to the occurrence and development of chronic diseases of young and middle-aged people,and the use of artificial intelligence methods for early warning of diseases should be promoted.,Only by cultivating multi-disciplinary interdisciplinary people can promote the combination of machine learning and health problems,and promote efficient health management. |
PDF Full Text Request