FMECA Of The Core Components Of Medical Accelerator Based On Text Mining

Reliability is an important index to measure whether the medical equipment is consistent with the quality management standard of medical device production.As large radiotherapy equipment,medical linear accelerator has strict requirements for reliability.The foreign radiotherapy equipment companies have had a set of perfect reliability implementation system and it has become a monopolistic factor of the radiotherapy market.The lack of engineering technology,weak basics and the deficiency of standard specification are the reliability consideration status of the radiotherapy equipment in China.Therefore,it is urgent to carry out the reliability research of the medical electronic linear accelerator in China.The accelerating tube is not only the core part but also the vulnerable part of the medical electron linear accelerator.Its main function is the output of the high-energy radiation therapy.The reliability research of the accelerating tube is the foundation of the reliability service of the medical electron linear accelerator.Failure mode,effects and criticality analysis(FMECA)is a commonly used method of reliability analysis.It can improve the reliability of products by finding out the weaknesses and improving them.Therefore,FMECA can be used to analyze the reliability of accelerating pipes.However,the large number of texts makes it's hard to carry out FMECA by a conventional way.In view of such situation,this paper proposes a method combining text mining and FMECA.The research in this thesis can be concluded as follows:Firstly,the function and structure of the accelerating tube are analyzed,and the agreement level of the acceleration tube is divided.Then,the combination of text mining and FMECA found out.And a kind of FMECA procedure is proposed based on text mining,and the corresponding model selection and parameters optimization are given.Additionally,the fault mode,fault cause and fault location information of the accelerated tube are extracted from the collected text data by text mining method.Three criteria are established,the failure mode probability rating score criterion,the fault pattern detection score criterion,and the severity level score criterion are established.Then the rating of the fault pattern and the corresponding hazard index are obtained using the extracted fault information and scoring criteriaFinally,the improvement measures and compensation methods are given for each type of failure mode,and the FMECA form and the list of key failure modes are output.The case of accelerating tube FMECA achieves the expected effect.It proves the feasibility of the combination of text mining and FMECA analysis.The research in this thesis provides a feasible way for the reliability analysis in big data,and can be used as reference for the reliability research of the acceleration tube system and the medical electronic linear accelerator.