SCIENCE@home

Near-Infrared based Non-Invasive Blood Glucose Monitoring using Statistical Models

S. Hemalatha,Dharun Muthaiah Nataraj,Hariharan Vijay Arun,Jagannath Mohan

Published 2025 in 2025 AI-Driven Smart Healthcare for Society 5.0

ABSTRACT

Diabetes management requires effective glucose monitoring, but most current techniques are intrusive, making patients uncomfortable and reducing compliance. This paper presents a non-invasive blood glucose monitoring system that combines Arduino UNO, an LCD, a buzzer, and a unique near-infrared (NIR) technology sensor. The device determines blood glucose levels by examining the NIR absorption properties of blood glucose molecules. Invasive blood glucose data was collected before and after meals from 15 participants for day 1 and day 2. Each reading averaged over 10–15 sensor measurements to ensure dependability. Predictive models such as random forest regression, support vector regression (SVR), and polynomial regression were used for validation. The root mean square error (RMSE) and mean average error (MAE) results reveal that the random forest obtained the best scores of 6.599 and 4.798 for the consolidated 60 values of the predicted glucose level. The suggested NIR-based system provides a non-invasive and real-time glucose monitoring option, and the results show its accuracy and viability. This method reduces invasive operations and can be integrated into wearable devices for continuous monitoring.

PUBLICATION RECORD

CITATION MAP

EXTRACTION MAP

CLAIMS

  • No claims are published for this paper.

CONCEPTS

  • No concepts are published for this paper.

REFERENCES

Showing 1-29 of 29 references · Page 1 of 1

CITED BY