Fast and noninvasive electronic nose for sniffing out COVID-19 based on exhaled breath-print recognition

The reverse transcription-quantitative polymerase chain reaction (RT-qPCR) is the standard for detecting SARS-CoV-2 but has limitations related to cost, invasiveness, turnaround time, and laboratory requirements. The study reports a low-cost, noninvasive, rapid screening method for COVID-19 using a portable electronic nose (GeNose C19) integrating an array of metal oxide semiconductor gas sensors with optimized feature extraction and machine learning models. In profiling tests using 615 breath samples (333 RT-qPCR-positive and 282 RT-qPCR-negative) collected from 83 patients in two hospitals in Yogyakarta, Indonesia, four machine learning algorithms (linear discriminant analysis, support vector machine, stacked multilayer perceptron, and deep neural network) achieved high performance, with detection accuracy of 88–96%, sensitivity of 86–94%, and specificity of 88–95% on testing datasets. The results suggest GeNose C19 is a promising breathalyzer for fast COVID-19 screening.