Non-invasive Blood Glucose Testing Based On Electrochemical Detection Technology

Diabetes is a chronic metabolic disease that causes many serious complications and has no obvious clinical manifestations in the early stages.Traditional invasive blood glucose testing not only causes discomfort to patients,but also increases the risk of cross-infection.Although the current research on non-invasive blood glucose testing has made some achievements,there are still some challenges that have not yet been overcome,for example,the instruments and equipment for optical and microwave testing are expensive,too large,and the testing time is long,which is not convenient for real-time monitoring of blood glucose.Compared with optical detection and microwave detection,electrochemical detection has the advantages of low cost,fast detection,simple operation.Therefore,this paper proposes a scheme to detect sweat glucose and thus indirectly detect blood glucose by electrochemical detection method to realize the non-invasive detection of human blood glucose.The main work is as follows:(1)To address the problem that sweat is volatile and difficult to collect during exercise,this paper designs a paper-based microfluidic sweat collection device based on microfluidic technology,using the core absorption characteristics of the paper base to liquids,to achieve efficient collection of sweat generated during exercise.(2)The GO-COOAu electrode and screen-printed gold electrode were constructed as enzyme-free glucose electrodes by chemical synthesis and screen-printing.The problem that the enzyme-free glucose electrode cannot detect glucose in neutral solutions is solved by raising the p H of the sweat by electrolysis or adding Na OH solution.A protective film was also modified on the surface of the working electrode to reduce the influence of other charged molecules in the sweat on the glucose detection.(3)Based on the electrochemical detection method,a sweat glucose hardware detection platform based on multi-step voltage step method was designed.The hardware platform mainly includes multi-step voltage selection module,potentiostat module and electrical signal conversion amplification module and other related hardware modules.First,the sweat p H value is stabilized to a specific value by applying negative potential electrolysis of sweat;After that,the set potential is applied to oxidize the glucose in the sweat,and the electrochemical signal is collected and processed by the electrical signal conversion amplification module after the oxidation is completed,and the glucose detection results are obtained.Finally,the positive potential is applied to remove the adsorbed oxide on the electrode surface to prevent interference with subsequent detection due to oxide adhesion.(4)In order to establish the relationship model between sweat glucose and blood glucose,this paper firstly analyses the feasibility of sweat glucose concentration to predict blood glucose concentration,and for the problem of large inter-individual variability,proposes to combine glucose data with four factors affecting metabolism as the characteristic parameters for blood glucose prediction.After that,the data processing method based on Dixon’s criterion,the blood glucose sample classification method based on SVM algorithm,and prediction model construction based on multiple linear regression,and the three methods are combined to build the blood glucose prediction model.Experimental validation was conducted on a non-invasive blood glucose detection system,and the results showed that: the average absolute error of blood glucose prediction results was about 0.57 m M,the prediction results have a certain reference,which can better reflect the trend of blood glucose changes in human body,and lay the foundation for the realization of non-invasive blood glucose testing.