论文部分内容阅读
利用人体血浆的表面增强拉曼光谱(SERS)并结合多变量统计方法对胃癌的无损诊断分析进行了研究.检测了32例胃癌患者与33例正常健康人血浆SERS光谱,利用主成分分析(PCA)并结合线性判别分析(LDA)建立SERS光谱诊断多元统计算法模型.为验证所构建的PCA-LDA算法的有效性,将利用受试样品的工作特征(ROC)曲线方法对所构建的算法有效性进行评价.胃癌患者与正常健康人血浆SERS光谱之间的差别明显,且实验存在较好的重现性,利用PCA-LDA统计分析方法得到诊断特异性与灵敏度分别为91%与79.5%.通过SERS谱峰归属分析表明,癌症患者血浆与正常人血浆在生化成分上存在一定的差异.与正常健康人相比,胃癌患者血浆中的核酸、胶原、磷脂以及苯丙氨酸成分偏高,而氨基酸与糖类成分相对偏低.研究表明,血浆SERS光谱技术结合PCA-LDA统计分析能够很好地区分正常健康人与胃癌患者血浆.血浆表面增强拉曼光谱技术有望发展为一种无损探测与筛查胃癌的临床诊断工具.
The plasma SERS spectra of 32 gastric cancer patients and 33 normal healthy individuals were detected by surface-enhanced Raman spectroscopy (SERS) combined with multivariable statistical methods.The PCA ) Combined with linear discriminant analysis (LDA) to establish a SERS spectral multivariate statistical algorithm model.In order to verify the validity of the constructed PCA-LDA algorithm, using the working characteristic (ROC) curves of the tested samples, the constructed algorithm The difference between the plasma SERS spectra of gastric cancer patients and normal healthy people was obvious, and there was a good reproducibility of the experiment.The diagnostic specificity and sensitivity of PCA-LDA method were 91% and 79.5% .Analysis of the SERS peak assignments showed that the plasma biochemical composition of plasma of cancer patients was different from that of normal individuals.Compared with normal healthy subjects, the plasma concentrations of nucleic acid, collagen, phospholipid and phenylalanine in gastric cancer patients were higher , While the amino acids and carbohydrate composition is relatively low.Studies show that plasma SERS spectroscopy combined with PCA-LDA statistical analysis can be very good distinction between normal and gastric cancer patients with plasma Plasma surface enhanced Raman spectroscopy is expected to develop as a clinical diagnostic tool for noninvasive detection and screening of gastric cancer.