基于最小二乘支持向量机的纳米金免疫层析试条快速定量

Quantitative Determination of Gold Immunochromatographic Assay based on Least Squares Support Vector Machine

通过光电反射式的光路扫描纳米金免疫层析试条测试线和质控线信号,研究基于最小二乘支持向量机的纳米金免疫层析试条快速定量方法,建立遗传算法优化的最小二乘支持向量机纳米金免疫层析试条定量研究方法。该方法对纳米金免疫层析试条甲胎蛋白(AFP)检验样本的统计数据中,样本相对均方差RMSE为12.2%,实验结果表明:遗传算法优化的纳米金免疫层析试条最小二乘支持向量机定量拟合模型有较好的整体性能和局部性能,适用于纳米金免疫层析试条的快速定量。

The least squares support vector machine （LSSVM） was applied to build the fitting model for quanti tative determination of nanogold immunochromatographic assay （GICA） strip based on the reflective optical detec tion. The genetic algorithm （GA） was used to solve the optimization problem of the I_SSVM model between the char acteristic parameters and the sample concentration. In the statistical data of the alphafetoprotein （AFP） GICA strip test samples ,the sample relative mean square error was 12.2%. The experimental results indicated that the least squares support vector machine model which optimizing by the genetic algorithm performed well, and proved to be appropriate in quantitative determination of GICA strip.