【摘要】：良好的血糖監(jiān)測能有效的診斷和預(yù)防糖尿病的發(fā)生,延緩糖尿病急、慢性并發(fā)癥的發(fā)展,而糖化血清蛋白的檢測能夠反映近期內(nèi)血糖控制水平。本論文采用二維相關(guān)光譜和密度泛函理論對糖化血清蛋白的光譜特異性進(jìn)行了研究,確定了糖化血清蛋白的特征吸收峰位置,并結(jié)合化學(xué)計量學(xué)方法建立了糖化血清蛋白濃度的定量校正模型,用驗證集對模型的預(yù)測能力進(jìn)行了檢驗。主要研究工作如下:(1)采用二維相關(guān)光譜分析和密度泛函理論計算,從實驗和理論兩方面分別分析了葡萄糖、血清蛋白、糖化血清蛋白的光譜,并對其吸收峰進(jìn)行了歸屬,確定了糖化血清蛋白的特征結(jié)構(gòu)N-H基團(tuán)振動的吸收峰在7053cm-1附近。在密度泛函理論計算中,對糖化血清蛋白相比于葡萄糖、血清蛋白偏差較大的吸收峰進(jìn)行歸屬,得到的三個基頻吸收峰2953.91 cm-1 3101.44cm-1 3528.14cm-1,倍頻后與二維相關(guān)光譜分析得到的倍頻吸收峰 5923cm-1 6359 cm-1、7053cm-1 相一致。(2)在確定了糖化血清蛋白的特異性以及特征吸收峰的位置后,采用遺傳算法-區(qū)間偏最小二乘法(GA-iPLS)建立糖化血清蛋白濃度的定量校正模型,其決定系數(shù)R2為0.9763,校正集樣本的均方根誤差RMSEC為0.007,遺傳算法優(yōu)選出的波段包含二維相關(guān)光譜分析和密度泛函理論計算得到的吸收峰,說明優(yōu)選的波段能很好的用于糖化血清蛋白濃度模型的建立。使用驗證集對模型的預(yù)測能力進(jìn)行了檢驗,預(yù)測模型的決定系數(shù)R2為0.9705,驗證集樣本的均方根誤差RMSEP為0.0087。結(jié)果表明,使用遺傳算法-區(qū)間偏最小二乘法建立的糖化血清蛋白近紅外光譜的濃度模型可用于糖化血清蛋白的臨床檢測。
[Abstract]:Good blood glucose monitoring can effectively diagnose and prevent the occurrence of diabetes, delay the development of acute and chronic complications of diabetes, and the detection of glycosylated serum protein can reflect the level of blood glucose control in the near future. In this paper, the spectral specificity of glycosylated serum protein was studied by two-dimensional correlation spectroscopy and density functional theory, and the position of characteristic absorption peak of glycosylated serum protein was determined. The quantitative correction model of glycosylated serum protein concentration was established by chemometrics method, and the prediction ability of the model was tested by verification set. The main research work is as follows: (1) the spectra of glucose, serum protein and glycosylated serum protein were analyzed by two-dimensional correlation spectrum analysis and density functional theory, respectively, and the absorption peaks of glucose, serum protein and glycosylated serum protein were assigned. It was determined that the absorption peak of N 鈮，

本文編號：2505335