Publication

Abstract : Diabetes has become a grave concern which can affect anyone irrespective of their age. To prevent the uneasiness of invasive glucose monitoring system, it is desirable to develop a low cost non-invasive blood glucose monitoring system. The present work proposes the feasibility of developing such a method for the continuous monitoring of blood glucose concentration using near infrared spectroscopy. Both linear regression and polynomial regression analysis are studied for developing an enhanced algorithm for estimation of glucose concentration using the scattering property of glucose molecules and the principle of photoplethysmography. Processing of data conversion is performed with MSP430G2553 microcontroller in the integrated development environment (IDE) of Code Composer Studio (CCS). Data analysis is carried out in MATLAB from which a relation between blood glucose concentration and photoplethysmograph is derived.