Publication

Abstract : Diabetes is one among the supreme health challenges of the current century. Most common method for estimation of blood glucose concentration is using glucose meter. The process involves pricking the finger and extracting the blood along with chemical analysis being done with the help of disposable test strips. Non-invasive method for glucose estimation promotes regular testing, adequate control and reduction in health care cost. The proposed method makes use of a near infrared sensor for determination of blood glucose. Near-infrared (NIR) is sent through the fingertip, before and after blocking the blood flow by making use of a principle called occlusion. By analyzing the variation in voltages received after reflection in both the cases with the dataset, the current diabetic condition as well as the approximate glucose level of the individual is predicted. The results obtained are being validated with glucose meter readings and statistical analysis of the readings where done. Analysis shows that the bias as well as the standard deviation decreases as the glucose concentration increases. The obtained result is then communicated with a smart phone through Bluetooth for further communication with the doctor.