Determination of Glucose Level in Blood and Aqueous Solution using Fourier Transform Near Infrared Spectroscopy

Abstract

Fourier transform near infrared spectroscopy is frequently utilized because it can measure a variety of solid and liquid samples, including components that are water soluble. The measurements of glucose concentrations in aqueous solutions are helpful to explore how near-infrared spectroscopy can be used for non-invasive assessments of glucose levels in the blood and how low concentrations of glucose in water may be determined using it. This study uses Fourier transform near-infrared spectroscopy to present an alternate method for estimating the glucose concentration in aqueous solutions below 1000 mg/dL. This method benefits from being non-destructive and less labor-intensive sample preparation. We carefully produced aqueous solutions of glucose at concentrations of 0–100 mg/dL at intervals of 5 mg/dL, 110–500 mg/dL at intervals of 10 mg/dL, and 525–1000 mg/dL at intervals of 25 mg/dL. As a result, 81 standard solution samples overall are produced for the calibration and validation sample sets. Inferring that the near-infrared prediction model is enough to estimate glucose content in the aqueous solutions below 1000 mg/dL, PLSR analysis to near-infrared spectra demonstrates that glucose content in aqueous solutions can be predicted effectively with a maximum variance of 6 mg/dL. While building a non-destructive glucose level measurement system employing a near-infrared light source, the NIR's capacity to detect glucose concentration below 1000 mg/dL is very crucial.