Intelligent Medical Image-assisted Diagnosis System Research On Index System

[Purpose]The combination of artificial intelligence and medical imaging is one of the most promising fields at present.It can deeply dig rich human health information in medical imaging data to help doctors make medical diagnosis.The standardized system evaluation index system can play an important role,but at present,China’s evaluation research on the system is relatively scarce.Existing research focuses on image quality evaluation and ignores the development of intelligent medical imaging assistance systems.Therefore,this research aims to build a set of evaluation index system suitable for China’s intelligent medical imageassisted diagnosis system,and promotes the standardization and scientification of the system by using this evaluation index system.[Methods] This study mainly analyzes the current status of relevant research at home and abroad through literature analysis,selects relevant indicators based on the analysis content and related principles of system,science,and practicality,and establishes a preliminary framework;and then uses the Delphi method to improve and modify After improvement,assign values according to the importance to the score and use the analytic hierarchy process;then design relevant questionnaires,survey specific populations,and collect evaluation data;use the reliability and validity of spss to test the reliability and effectiveness of the index system To test the degree of reliability of the indicator.[Results] This study constructed a primary evaluation index system from the four dimensions of technical level,system function,system performance and system application.The index system finally obtained 4 first-level indicators,16 second-level indicators and 58third-level indicators.And give weights by AHP.Finally,a questionnaire was designed based on the indicator system and the questionnaire survey was conducted on the personnel of the relevant institutions.The reliability and validity of the indicator system were tested by statistical relevant testing methods.It is found that there are still problems such as inconsistent standards,imperfect systems,image data sharing and accuracy.[Conclusion] This study applies the Delphi expert method combined with analytic hierarchy process and other methods to the index system,and evaluates the intelligent imageassisted diagnosis system from multiple aspects.The analysis of experts and questionnaire survey data found that the evaluation results of this study are consistent,which shows that the evaluation index system constructed by this research is scientific and reasonable.