Design And Implementation Of Web-based Intelligent Pre-diagnosis System

As the contradiction between the unbalanced social medical resources and the increasing health care needs of the people is getting worse,pre-examination is gradually becoming popular as a significant method to improve the efficiency of medical treatment.This article will take ophthalmology as an example to conduct research in the field of task-based dialogues for the purpose of pre-diagnosis,and convert the results into an intelligent Web-based pre-diagnosis system.This article first uses text structuring technology to extract ophthalmic disease feature fields on 58 health and other websites and build a disease knowledge base.Then,based on the disease prediction strategy,it has achieved multi-round dialogue to collect user disease information,and targeted Carry out questioning and clarification,and then convert the available information into pre-diagnosis reports for storage,and then integrate the intelligent pre-diagnosis and the feedback of its results into an intelligent pre-diagnosis system.Finally,the relevant research on the interface configuration of the dialogue robot is carried out,A one-stop assistance system for training and testing the functions of the department's assistants.Based on medical text structuring,knowledge base construction,natural language analysis,dialogue management,disease prediction analysis,text generation and system construction for pre-diagnosis reports,this paper mainly completes the following work:(1)Proposing and designing intelligent pre-examination assistants.Taking ophthalmology as an example,through the research on the content and process of the inquiry,the human-machine inquiry dialogue is designed in an external series and in parallel manner,and the algorithm engine is designed,including a multi-round dialogue module,an NL2SQL module,and a proper noun matching module.A set of training data corresponding to it.In addition,this article also proposes two disease prediction strategies,which are selecting the most specific symptom questioning and selecting the most frequent symptom questioning,to lay the foundation for user feedback.(2)Based on this intelligent assistant,a Vue-based intelligent pre-examination system is designed and implemented.The system includes four major functions of pre-examination dialogue,department assistant management,dialogue history management,and personal health management.Based on the patient information collected by the ophthalmic pre-examination assistant and an electronic pre-examination report for its disease prediction output specification,and Provide patients with pre-diagnostic services such as nursing advice and reminders.(3)Designed and implemented a configurable multi-round dialogue system that provides assistant creation services for the pre-examination system,including account management module,corpus management module,robot configuration module,human-computer interaction management module,and training visualization module.Through the open source machine learning framework Rasa based on automatic text and voice-based dialogue,the visual process orchestration of robot application construction was carried out in order to reduce the threshold for the development of dialogue systems.Based on the above research content,from the practical perspective of doctors and patients,this article studies the consultation process using ophthalmology as an example.Based on multi-round dialogue technology,an intelligent pre-examination system that supports user guidance,predicts diseases,and explains professional vocabulary is implemented for patients.Users provide access to intelligent pre-examination and feedback services.In response to the high cost of intelligent assistant development,we have developed a configurable multi-round dialog robot system to provide an interface-based one-stop intelligent assistant creation service for the intelligent pre-examination system or other fields of engineering.Two patents have been published based on the research content.