Exhaled breath-print analysis by using metal oxide chemoresistive sensors for classifying and monitoring patients with different clinical states

Health status monitoring based on non-invasive methodology through exhaled breath analysis has raised great interest, due to its easiness that does not require skilled medical personnel. The aim of the present study is to discriminate between exhaled breath samples of patients' groups with Diabetes Mellitus (DM), Renal Failure (RF), Liver Cirrhosis (LCi), and Healthy Controls (HC), by using an array of metal oxide chemoresistive sensors. Breath samples collected from HC (n=10), DM (n=6), RF (n=11), and LCi (n=11) patients were analyzed by the electronic nose (e-nose), and classification was performed using chemometric techniques namely: Discriminant Function Analysis (DFA) and Support Vector Machines (SVMs). As result, DFA has shown good discrimination between data-points of breath samples related to HC, DM, RF, and LCi patients. The SVMs method reached a 96.49% success rate for the recognition of the analyzed four groups. In the light of these results, we can state that the presented e-nose technology demonstrates that a simple, cost-effective, and non-invasive approach based on exhaled breath analysis could be considered a reliable screening tool for diseases diagnosis.