中文摘要：

本文提出了一种面向冠心病血检数据的函数型主成分分析（FDCA）方法。首先,利用函数型数据分析（FDA）方法,对冠心病患者的血液检测指标数据进行函数化;然后,利用FPCA的方法对原始血检指标进行降维,并与传统的主成分（PCA）方法进行对比。实验结果表明,相比于传统方法,FPCA方法降维效果更好,降维后得到的新指标能够包含原始指标的绝大部分信息。该方法对于临床冠心病的诊断研究具有现实和理论意义。

英文摘要：

This paper presents a new method of functional data principal component analysis for coronary heart disease （CHD）. First, the data of CHD blood test are being functional and continuous by using the principle of functional data analysis. Then, we reduce the dimensions of the original test index by introducing the method of functional principal components analysis （FPCA） and compare the corresponding results with the cases of the methods with principal component analysis （PCA）. The results show that the method of FPCA is more effective, and the new indicators, obtained by dimension reduction, can contain most of the original index information. This will help the clinical diagnosist for CHD.