Validation of Analytical Models for the Development of Non-Invasive Glucose Measurement Devices

Area of Science:

• Biomedical Optics
• Analytical Chemistry
• Spectroscopy

Background:

• Non-invasive glucose monitoring is a significant challenge.
• Traditional optical methods and direct Lambert-Beer Law application face limitations with biological samples.
• Near-infrared (NIR) spectrophotometry shows promise but requires refined methodologies for complex matrices.

Purpose of the Study:

• To propose and validate a methodology for determining glucose extinction coefficients in blood.
• To address the limitations of the Lambert-Beer Law in biological samples.
• To account for molecular interactions and non-linear behaviors in spectrophotometric analysis.

Main Methods:

• Spectrophotometry was used to analyze whole blood and plasma samples (700-1400 nm).
• An iterative optimization process was employed to determine glucose extinction coefficients.
• A regression model was developed considering cuvette thickness and absorbance at specific wavelengths (965, 975, 985 nm).

Main Results:

• Glucose exhibited high spectral sensitivity near 975 nm.
• Obtained extinction coefficients for glucose (αg) ranged from -0.0045 to -0.0053.
• A regression model successfully differentiated glucose from insulin concentrations, with stable parameters observed.

Conclusions:

• A wavelength of approximately 975 nm is optimal for blood glucose calculation via photometry.
• The proposed model accounts for glucose behavior in biological media, aligning with mid-infrared validation studies.
• The methodology offers a pathway to overcome Lambert-Beer Law challenges in non-invasive glucose monitoring.