摘要当今人类的生活水平不断提高，进食过多，体力活动减少以及各种遗传因素，使得糖尿病成了当前威胁全球人类健康的最主要的慢性非传染性疾病之一。目前临床或家庭的血糖监控多是采用人体血液取样的试纸检测方法，这种具有创伤的血糖常规检测，会给糖尿病患者带来痛苦，因此本文研究使用近红外光谱法无创检测血糖。 本文选择用葡萄糖溶液替代房水样品溶液，通过对葡萄糖溶液中葡萄糖浓度的测定来模拟对房水中血糖浓度的测定。在配置溶液的过程中尽可能的减少误差以保证最终结果的准确性。利用近红外分光光度计采集葡萄糖溶液的透射光谱，使用偏最小二乘法建立校正模型，并且对葡萄糖溶液样本浓度进行预测。用预测标准差(RMSEP)来表示预测结果的准确度。 通过对浓度分别为20-340mg/dL和500-10000mg/dL的葡萄糖溶液吸光度的检测结果分析，可以得出如下结论：本文使用的算法在低浓度情况下得到的结果并不是很理想；但在高浓度情况下可以很好的用于血糖信号的数据处理，证明了校正模型有良好的预测能力，所以应用前景还是很广阔的。 51317
毕业论文关键词 血糖无创检测 近红外光谱 血糖预测 房水
Title Near infrared spectroscopy signal analysis of blood glucose simulated sample
Abstract Nowadays, people’s living standards continue to improve. Eating too much, decreased physical activity and variety of genetic factors make diabetes one of the main chronic non communicable diseases which make global people under threat. Current clinical or family blood glucose monitoring most uses the method of human blood sampling test strip detection. The traumatic blood glucose routine testing will bring pain to patients with diabetes. In this paper, aqueous humor sample solution was replaced by glucose solution. The simulation test of blood glucose concentration in aqueous humor is through detecting the concentration of glucose in glucose solution. In order to ensure the accuracy of the final results, errors should be reduced as much as possible in the process of configuring solution. The transmission spectrum of glucose solution was collected by near-infrared spectrophotometer. The calibration model was established by partial least square(PLS) method so that the concentration of glucose solution was predicted. The Root Mean Square Error of Prediction (RMSEP) is used to show the accuracy of the prediction results. According to the analysis of the detection results of absorbance of glucose solution with concentration of 20-340mg/dL and 500-10000mg/dL, the conclusion is drawn as follows: The results obtained in this paper are not very good at low concentration. However, in the case of high concentration algorithm can be well used in blood glucose signal data processing. It proves that the calibration model has good predictive ability. So the application prospects are very broad.
Keywords Glucose non-invasive detection Near Infrared Spectroscopy Blood glucose prediction aqueous humor
目次
1 引言 1
1.1 课题的研究背景及研究意义1
1.3 本文研究内容3
2 近红外光谱无创检测血糖的光谱特性 5
2.1 分子吸收光谱 5
2.2 红外光谱的产生条件 5
2.3 分子振动方程及分子的振动形式 6
3 近红外光谱无创检测血糖的光路设计8
3.1 人眼结构 8
3.2 近红外光谱检测光路方案 8 血糖模拟样品的近红外光谱信号分析 :http://www.youerw.com/tongxin/lunwen_54916.html