DESIGN AND OPTIMIZATION OF GAS SENSORS FOR DETECTION OF VOC BIOMARKERS IN BREATH

This thesis explores the development and optimization of gas sensors for detecting volatile organic compounds (VOCs) in breath, with a focus on Tin Oxide (SnO₂) metal oxide sensor arrays. Simulated breath experiments were conducted to evaluate the sensors' capabilities in a controlled environment, determining their ability and efficacy in detecting key VOCs linked to various health conditions. The sensors were able to detect even minor fluctuations in exhaled breath compounds, making them excellent candidates for continuous and non-invasive health monitoring applications. Significant correlations were identified within sensors and between specific extracted features, facilitating the optimization of feature selection for further analysis. For the analysis of human breath, a total of 164 samples were collected and analyzed from different individuals, with ten providing nine additional samples for longitudinal assessment. Analysis of data from different subjects revealed confounding variables such as age and gender showed minimal impact on sensor performance. Analysis of the longitudinal data showed that variability between subjects was significantly higher than within replicates of a single volunteer. Multivariate analyses displayed that subjects could not be distinguished from one another. VOC data from the sensor array can be cross-referenced in future studies aiming to use the device to distinguish disease states. Ultimately, the MOX sensors along with sensors specific to different analytes may be integrated into a portable breathalyzer for rapid and non-invasive health monitoring.