AI Based Non-invasive Glucose Detection Using Urine

This proposed device uses urine to predict the glucose level present in the patient using non-invasive technique with a high level of accuracy for detection of diabetes. The paper presents a urine glucose level diagnosing and prediction using a computer-based polarimeter held in a portable device, to provide a fast and accurate on-field result. The instrument consists of an LCD screen, optical sensor, Benedict’s reagent, a detachable tank, and an embedded system-on-chip (SoC).