论文部分内容阅读
使用激光共聚焦拉曼光谱仪测量正常大鼠红细胞、正常人红细胞、糖尿病STZ造模大鼠红细胞、糖尿病四氧嘧啶造模大鼠红细胞和人Ⅱ型糖尿病红细胞的拉曼光谱,应用主成分分析(principal component analysis,PCA)结合支持向量机(support vector machines,SVM)分类器对数据进行判别分析,然后采用类间距离判断两种造模方法与人Ⅱ型糖尿病的接近程度。结果发现糖尿病红细胞与正常红细胞的拉曼光谱存在明显差异,糖尿病在酰胺ⅥCO变形振动谱带处峰高显著,并在酰胺ⅤN—H变形振动谱带处谱线出现偏移,属于磷脂的脂酰基C—C骨架1 130cm-1谱线增强,1 088cm-1谱线强度减弱,说明糖尿病红细胞膜的通透性增强。PCA结合SVM可以很好地区分以上5类红细胞的拉曼光谱,分类器测试结果表明分类准确度达100%。通过分别计算两种造模方法与人Ⅱ型糖尿病的类间距离,发现STZ造模法更接近人Ⅱ型糖尿病。由此得出结论:拉曼光谱法可以用于糖尿病诊断,大鼠糖尿病STZ造模法更接近人类Ⅱ型糖尿病。
Laser Raman spectroscopy was used to measure the Raman spectra of erythrocytes of normal rats, normal human erythrocytes, erythrocytes of diabetic STZ-model rats, erythrocytes of diabetic alloxan-induced diabetic rats and human type II diabetic red blood cells. Principal component analysis principal component analysis (PCA) and discriminant analysis based on support vector machines (SVM) classifier, and then use the distance between classes to determine the closeness of the two methods to human type II diabetes. The results showed that the Raman spectra of diabetic erythrocytes and normal erythrocytes were significantly different. Diabetes mellitus had a prominent peak at the Ⅵ C O deformation vibrational band, and shifted in the spectrum of the ⅤN-H deformation vibrational spectrum. The acyl C-C backbone 1 130 cm-1 enhanced and the intensity of 1 088 cm-1 decreased, which indicated that the permeability of diabetic erythrocyte membrane increased. PCA combined with SVM can well distinguish the above five types of red blood cell Raman spectra, classifier test results show that the classification accuracy of 100%. By separately calculating the distance between two modeling methods and human type II diabetes, it was found that the STZ modeling method is closer to human type II diabetes. From this conclusion: Raman spectroscopy can be used for the diagnosis of diabetes, rat diabetes STZ modeling closer to human type II diabetes.