Breath Analysis Using Quartz Tuning Forks for Predicting Blood Glucose Levels Using Artificial Neural Networks

Area of Science:

• Biomedical Engineering
• Analytical Chemistry
• Computational Biology

Background:

• Diabetes Mellitus (DM) is a prevalent metabolic disorder requiring continuous monitoring.
• Current blood glucose tests are invasive and necessitate medical intervention.
• Development of noninvasive diagnostic tools for DM is crucial for effective disease management.

Purpose of the Study:

• To introduce a novel noninvasive method for diabetes diagnosis and monitoring using breath analysis.
• To evaluate the efficacy of nanomaterial-enhanced quartz tuning fork sensors coupled with artificial neural networks (ANN) for DM detection.
• To assess the accuracy of ANN-based blood glucose prediction compared to traditional methods.

Main Methods:

• Utilized a sensor array of 12 nanomaterial-enhanced quartz tuning fork sensors to capture breath signatures.
• Employed customized artificial neural network (ANN) classification algorithms for data interpretation.
• Developed an ANN regression algorithm to predict blood glucose levels and validated results using error grids.

Main Results:

• The sensor array and ANN system achieved 97% accuracy in identifying diabetic, prediabetic, and healthy individuals.
• ANN regression predicted blood glucose with a correlation coefficient of 0.89 and a mean square error of 0.13.
• The noninvasive method demonstrated high clinical relevance in assessing blood glucose values.

Conclusions:

• Breath analysis using nanomaterial-enhanced sensors and ANN provides a highly accurate, noninvasive approach for DM detection.
• This technology offers a promising alternative for convenient and frequent monitoring of diabetes.
• Further research can explore the integration of this system for widespread clinical application.