中文摘要：

研究了扩散加权的NMR在区分冠心病患者和正常人的血清中的可能应用,证明了扩散加权的NMR与主成分分析（PCA）相结合是将冠心病血清从正常血清中识别出来的有效方法。分子自扩散运动加权的NMR能够有效抑制小分子代谢物的共振信号强度,从而使血脂和血蛋白的NMR特征更为明显。最佳的扩散权重因子（b）约为0.85×10^7s/cm^2,增加或降低扩散权重因子都会影响识别效率。研究结果表明,分子量极小的代谢物（乳酸等）及相对较大的脂蛋白（VLDL）对区分两类血清的贡献较小。在最佳的b值条件下,高密度脂蛋白（HDL）和低密度脂蛋白（LDL）对区分两类血清的贡献得以提高,从而获得更好的区分效果。

英文摘要：

It is demonstrated that combination of diffusion-weighted NMR spectroscopy and principle components analysis（PCA） can be used to discriminate serums of patients with coronary heart disease（CHD） from that of health subjects.Diffusion-weighted 1H NMR spectroscopy is a powerful approach for enhancing resonances from large molecules and suppressing resonances of low molecular-weight metabolites without disturbing physiological properties of human serum samples.The best diffusion weighting factor is found to be around b=0.85×10^7 s/cm^2,increasing or decreasing the diffusion weighting factor destructs the discrimination efficiency.These preliminary results show that high density and low density lipoproteins（HDL ＆ LDL） are the key components contributed to the discrimination,while very small metabolites（such as lactate etc.） and larger lipoproteins（such as very lower density lipoproteins,very low density lipoproteins（VLDL）,etc.） have less effects.