Smartphone-based colorimetric detection to measure Blood Glucose Levels

Abstract

Diabetes Mellitus is a hazardous disease plaguing many humans in the world and it is caused due to high Blood Glucose Levels (BGL). Patients with diabetes need to regularly check their BGL in order to stay safe. Through this paper, we aim to provide a medical image processing system based on colorimetric detection particularly for diabetic patients that will measure the blood glucose levels by scanning an image of the visual glucose test strip using a smart-phone. The proposed technique measures the color change of the scanned test strip by using HSV and CIELAB color model based processing wherein concepts of color-gradient and CIEDE2000 color-difference formulas are applied. The technique was implemented in an Android based environment and tested using Accu-Chek active visual test strips along with its corresponding reference color chart. The results of the experiment show that such technique has promising capability to measure BGL in humans. Further, such technique can be also applied for any other visual medical test strips or even for other measurements which are calculated using a color metric.