Non-invasive blood glucose measurement scheme based on near-infrared spectroscopy

Diabetes is a kind of common chronic diseases, seriously harmful to human health. Determination of blood glucose levels is the key link in diabetes care. The traditional measurement method firstly collects blood drops, and then uses the blood analyzer for electrochemical analysis. The traditional method is inconvenient, algesiogenic and easily cause infection. With the development of chemometric methods, infrared spectroscopy has been widely used in the field of blood glucose measurement and medicine care. The non-invasive blood glucose measurement scheme based on near-infrared spectroscopy is a way which is more efficiently, without damage and chemical reagents. In this paper, we proposed a non-invasive blood glucose detection and analysis the scheme based on near-infrared spectroscopy. In order to avoid the influence of the environment during the measuring process and considering the individual differences, we use the wave acquisition system to collect human pulse wave, and reduce the noise of the pulse wave signal first. By comparing the actual measured data with the data obtained in the traditional way, we can establish models to predict the blood glucose levels and achieve non-invasive blood glucose measurement using spectrum analysis technology. In the end, we improve the models by analyzing the relationship between individual factors and the pulse wave signal. The experimental results show that the method can effectively determine the blood glucose levels, and it is more convenient, safer and healthier than the traditional method, which can effectively reduce the time cost and the money cost.