Machine Learning Method and Hyperspectral Imaging for Precise Determination of Glucose and Silicon Levels

Area of Science:

• Analytical Chemistry
• Biomedical Engineering
• Artificial Intelligence

Background:

• Accurate glucose monitoring is vital for diabetes management.
• Silicon plays a key role in bone health, collagen formation, and connective tissues.
• Understanding silicon's physiological roles is essential for comprehensive health assessment.

Purpose of the Study:

• To introduce an algorithm for simultaneous detection of glucose and silicon levels.
• To leverage hyperspectral data and AI for enhanced monitoring.
• To evaluate the efficacy of Support Vector Machine Regression (SVMR) and perceptron models.

Main Methods:

• Integration of hyperspectral imaging and artificial intelligence.
• Development and comparison of SVMR and perceptron models for regression.
• Feature selection, model training, and rigorous evaluation metrics were employed.

Main Results:

• The algorithm demonstrated high accuracy in predicting glucose and silicon concentrations.
• Both SVMR and perceptron models showed effectiveness in the detection process.
• The study validates the potential for real-world application in monitoring systems.

Conclusions:

• The proposed AI-driven algorithm offers a reliable method for simultaneous glucose and silicon level detection.
• This technology has significant potential for advancing continuous monitoring systems.
• Further research can explore applications in personalized healthcare and disease management.