Sensor systems

Abstract

This chapter introduces the topic of detection of volatile organic compounds (VOCs) via nanomaterial-based sensors for health applications. Sensors are fast and relatively inexpensive tools for noninvasive diagnostics of health status by the detection of VOCs excreted from human breath or different body fluids and tissues, and they can be used to determine the volatilomic print of an individual. Utilization of the knowledge on different chemical families toward tuning sensor approaches is demonstrated. The chapter introduces the approach of semiselective/cross-reactive sensors for VOC detection and its advantages compared to selective sensor approaches. Different sensor technologies are reviewed, and it is shown that they are capable of easy adaptation in the field. As sensor systems provide high-dimensional outputs when applied to volatilomic analysis, predictive models for diagnosis are an integral part of the analysis. The chapter reviews current developments and future advancements of nanomaterial-based sensors and assesses the hidden difficulties in developing machine learning models for volatilome applications, offering recommended best practices for model validation and assessment.