Exhaled breath analysis using electronic nose and gas chromatography–mass spectrometry for non-invasive diagnosis of chronic kidney disease, diabetes mellitus and healthy subjects

• Published in: Sensors and actuators. B, Chemical Vol. 257; pp. 178 - 188
• Main Authors: Saidi, Tarik, Zaim, Omar, Moufid, Mohammed, El Bari, Nezha, Ionescu, Radu, Bouchikhi, Benachir
• Format: Journal Article
• Language: English
• Published: Lausanne Elsevier B.V 01.03.2018; Elsevier Science Ltd
• Subjects: Breath analysis; Breath tests; Chemical sensors; Chemometrics; Chromatography; Chronic kidney diseases; Cluster analysis; Creatinine; Diabetes; Diabetes mellitus; Diagnosis; Electronic nose; Electronic noses; Gas chromatography; Gas chromatography quadrupole time-of-flight mass spectrometry; Gas sensors; Kidney diseases; Mass spectrometry; Medical diagnosis; Organic chemistry; Pattern recognition; Principal components analysis; Quadrupoles; Regression analysis; Regression models; Spectrophotometry; Spectroscopy; Support vector machines; Urine; VOCs; Volatile organic compounds; Gas chromatography quadrupole time-of-flight mass spectrometry; Diabetes mellitus; Chemometrics; Breath analysis; Electronic nose; Chronic kidney diseases
• ISSN: 0925-4005; 1873-3077
• DOI: 10.1016/j.snb.2017.10.178

[Display omitted] •E-nose and GC/Q-TOF-MS were employed to investigate breath of CKD, DM and HS.•GC/Q-TOF-MS detected special breath VOCs with high concentrations for each case.•E-nose could characterize breath patterns of CKD, DM as well as HSHC and HSLC.•Eight new cases, i.e. 2 CKD, 2 DM, 2 HSHC and 2 HSLC were successfully identified.•Significant correlation was obtained between breath VOCs and urinary creatinine. Breath Volatile Organic Compounds (VOC’s) analysis is a non-invasive tool to assess information about health status. This study aims to investigate exhaled breath of Chronic Kidney Disease (CKD), Diabetes Mellitus (DM) and Healthy Subjects (HS), using electronic nose (e-nose) and Gas Chromatography Quadrupole Time-Of-Flight Mass spectrometry (GC/Q-TOF-MS). Breath samples were collected from 44 volunteers containing 14 females and 30 males. Urine samples were also collected to measure Creatinine Level (CL) by UV–vis Spectrophotometry as reference method. GC/Q-TOF-MS was used to identify volatile organic compounds that were detected in the exhaled breath of CKD, DM, and healthy subjects at different CL concentrations. The e-nose dataset was treated with Principal Component Analysis (PCA), Support Vector Machines (SVMs), Hierarchical Cluster Analysis (HCA) and Partial Least Squares-regression (PLS-regression). PLS model revealed a relationship between breath and urinary CL. The presented results show that e-nose based on chemical gas sensors in combination with pattern recognition methods could constitute the basis of inexpensive and non-invasive diagnosis to distinguish between breath of CKD, DM patients and healthy controls based on breath VOC’s analysis.