一种全自动酶免分析仪移液过程评估新方法  

New method of evaluating the pipetting of automatic ELISA system

作　　者：常海涛[1,2] 祝连庆[2] 娄小平[2] 郭阳宽[2] 王中宇[1] 

机构地区：[1]北京航空航天大学仪器科学与光电工程学院,北京100191 [2]北京信息科技大学光电测试技术北京市重点实验室,北京100192

出　　处：《仪器仪表学报》2014年第7期1622-1629,共8页Chinese Journal of Scientific Instrument

基　　金：北京市自然科学基金重点项目(KZ201010772032);教育部"长江学者与创新团队"发展计划(IRT1212);国防技术基础科研项目(J132012C001)资助

摘　　要：为了提高全自动酶免分析仪微量移液系统的可靠性,对移液过程压力监控曲线建立了基于灰色聚类分析的定性评估模型,对正常移液过程以及气泡、堵针、样本量不足等异常移液过程进行准确判定及报警。首先,利用差分法求取正常移液过程压力一阶导数曲线,以此作为参照建立移液过程的指标模型,并采用基于中心点三角白化权函数对多样性的评价指标进行灰量白化;然后,按照聚类指标将移液过程进行灰色聚类,计算不同移液过程所有指标的综合聚类效果,并以聚类类型为依据对移液过程进行定性评估。在此基础上,对压力监控一阶导数曲线进行灰色动态平滑滤波,消除基线漂移和随机噪声。最后搭建了微量移液实验系统,并对任取的11次移液过程进行定性综合评估。实验结果表明,该定性综合评估方法能够对正常及各种异常移液过程进行准确判定,并且给出问题类型。In order to improve the pipetting reliability of automatic enzyme-linked immunoassay analysis system, a new qualitative evaluation model based on grey clustering for the pressure monitoring curve of pipetting process is established, which can accurately perform judgment and warning for the normal pipetting process and abnormal pipetting processes with the pipetting errors, such as insufficient reagent, clogged tip and air bubbles. Firstly, with difference method the first order derivative curve of the pressure monitoring data in normal pipetting process is acquired, taking which as the reference the criteria model of the pipetting process is built. The center-point triangular whitenization weight functions are adopted to whiten the evaluation criteria with diversity. Secondly, all the pipetting processes are clustered with clustering criteria, the integrated clustering effects of all the criteria of various pipetting processes are calculated, and according to the clustering type the performances of the pipetting processes are qualitatively evaluated. On this basis, grey dynamic smooth filtering is performed on the first order derivative curve of the pressure monitoring data, which eliminates the baseline drift and random noise. Finally, a pipetting experimental system was built, 11 pipetting processes were selected arbitrarily, and the qualitative comprehensive evaluation was conducted. The experiment results indicate that the proposed model can accurately judge the normal and various abnormal pipetting processes and provide the error types.

关 键 词：移液过程定性评估 灰色聚类 灰色动态平滑滤波 压力监控 

分 类 号：TH773[机械工程—仪器科学与技术]