Design And Implementation Of A Distributed Auxiliary Detection System For Chronic Obstructive Pulmonary Disease

The diagnosis of traditional diseases mainly relies on doctors' consultation.This process is greatly influenced by the individual professional qualities of doctors.The goal of diagnosis about modern disease is to reduce the influence of subjective factors,and make it modernized,professional and intelligent.Therefore,establishing an auxiliary disease detection system already becomes the main trend of medical development.Chronic Obstructive Pulmonary Disease(COPD)is a chronic lung disease with difficulty breathing and long duration.It has the characteristics of high morbidity and high mortality.And has a serious impact on the patient's ability to work and quality of life.If the COPD patients can't timely find out their own physical condition and the development of the disease,it will have a great impact on the individual's psychology.Therefore,the establishment of auxiliary testing system for chronic obstructive pulmonary disease can not only pay attention to the health of COPD patients,but also improve the efficiency of doctors' consultation,and can also help patients to establish their confidence in rehabilitation.(4)The paper conducts an in-depth study on domestic and foreign auxiliary detection systems and summarizes the existing deficiencies of the current auxiliary detection system.This paper studies the main methods of system platform construction and analyzes the needs of the distributed auxiliary detection system for chronic obstructive pulmonary disease.And finally we decide to adopt the B/S architecture,the MVC design pattern and Java language to implement the system.The data of this system is distributed on HDFS,and data mining algorithms are distributed on Spark.The paper fully explains the design and implementation process of the distributed detection system for chronic obstructive pulmonary disease.The system mainly includes the following functions:(1)Using the MVC architecture to achieve the Multi-module functional integration,and mainly including six modules: health records,auxiliary diagnosis and treatment,appointment registration,verification code detection etc.This can assist COPD patients to detect the disease,simplify the procedure,and provide more information on chronic obstructive pulmonary disease.(2)Using the data mining algorithm to train the data in the medical records of434 patients with lung disease,which are collected from the Department of Pulmonology,Shandong Provincial Hospital of Traditional Chinese Medicine,and a classification model was established.The system provides auxiliary diagnosis and treatment for the patient's uploaded condition,diagnoses the patient's grade of COPD,and gives adjuvant treatment programs.(3)The system uses the verification code sliding detection method to perform login identity detection for users.In this paper,we use the mouse trajectory collected by the product,establish feature engineering,and propose a semi-supervised mouse trajectory recognition method based on parallel voting decision trees,which can effectively recognize machine trajectory.Only in this way can protect the security of the user account.The accuracy of the post-training disease detection model for identifying patients with COPD was 91%.The average accuracy for identifying the grade of COPD is60%;The AUC value of identifying human-machine mouse trajectories is 93%.