Development of Non-Contact Measurement of Glucose in Urine by Smartphone-Based Laser Refractometry for Continuous Monitoring of Diabetes

Amirul Badiuzzaman, Sinin and Khairul Fikri, Tamrin and Muhammad Hamdi, Mahmood (2025) Development of Non-Contact Measurement of Glucose in Urine by Smartphone-Based Laser Refractometry for Continuous Monitoring of Diabetes. Masters thesis, UNIMAS.

	PDF dsva_Amirul Badiuzzaman.pdf Restricted to Repository staff only Download (1MB) | Request a copy
	PDF (Please get the password by email to repository@unimas.my , or call ext: 3973 / 3942 / 3933) Thesis Master_Amirul Badiuzzaman.pdf Download (1MB)
	PDF Thesis Master_Amirul Badiuzzaman_24 pages.pdf Download (303kB)

Abstract

Diabetes mellitus is a global health issue affecting millions of people, requiring regular glucose level monitoring. Current non-invasive methods include urinalysis (colorimetry and biosensors) which are laboratory-based and lack user-friendliness, limiting their practicality for continuous glucose monitoring. Although promising, research on smartphone-integrated laser refractometry for glucose detection remains limited. To address this need, a non-contact smartphone-based laser refractometer for glucose monitoring was developed. This prototype measures the refractive indices of urine by analyzing the refracted length of a laser line, which is correlated with fasting blood glucose concentrations. The proposed prototype uses a smartphone to capture a relatively high resolution images of a laser line caused by total internal reflection through the rod and refraction caused by the urine. Assessments were made through a series of controlled glucose concentrations, varying turbidities, different volumetric samples, and a shelf-life. Volumetric and shelf-life assessments showed no effect to the results whereas as turbidity assessments proved the proposed prototype is limited to 57 nephelometric turbidity units (NTU) of the urine samples. Results of fasting glucose levels measured in urines using the developed system were compared to fasting blood glucose laboratory results, yielding with correlation coefficients of 0.89 and a sensitivity of 4.8 mg/dL. The system is inexpensive, making it accessible, and is suitable for telemedicine applications, providing remote monitoring options for patients. This approach paves the way for clinically crucial glucose detection in diabetics without the need for invasive finger-prick blood sampling.

Actions (For repository members only: login required)

View Item