Paper Title
Personalized Non-Invasive Blood Glucose Monitor

Abstract
Non-Invasive Blood Glucose Monitor is developed for 4 decades but it is still clinically unacceptable in the market due to its poor accuracy. The main reasons for poor accuracy are sensor accuracy, within-patient and in between patient variation. Most of the inventions are done to improve the non-invasive sensor accuracy. This paper presents, the concept of “personalized non-invasive blood glucose monitor” to improve the accuracy by resolving both within and in between patient variations. Our concept consists of both invasive and non-invasive sensors on a single device. Initially, the patient monitors his/her blood glucose level using both invasive and non-invasive approaches. The device processor calculates the error in non-invasive measurements with respect to invasive glucose value. Based on collected data, software develops an error function model by best suitable regression analysis to reduce the sum of square error. Once the device is calibrated for a patient, the error function model compensates for the error in non-invasive measurement and gives accurate non-invasive blood glucose results. At this point, the patient measures blood glucose levels by non-invasive methods only. The personalized blood glucose monitoring concept is applied and accuracy is improved from -221% ~ 61% to -14% ~ 18% for a dataset. Keywords - Blood Glucose Monitoring, Non-Invasive Blood Glucose Monitor, Personalized Glucose monitor, Non-Invasive Diabetic Care, Invasive Vs Non-Invasive Glucose Monitor.