Breath analysis is of great interests and widely needed for clinical practices, which ensures a non-invasive diagnosis of different health status. This study aims to investigate exhaled breath of Chronic Kidney Disease (CKD) patients during Hemodialysis (HD). To reach this aim, an electronic nose (e-nose), based on six SnO₂ sensors, coupled with principal components Analysis (PCA), Discriminant Function Analysis (DFA) and Support Vector Machines (SVMs), as alternative tools for the resolution of complex classification situations, were used. As results, the application of PCA, DFA and SVMs methods revealed a good separation between clusters of the breath samples of healthy subjects and CKD patients before, during and after HD. Consequently, the data treatment by DFA have demonstrated its efficiency to discriminate between CKD at pre-HD and post-HD. These findings indicate that the proposed e-nose could be a noninvasive diagnosis tool of breath analysis for real time monitoring of HD treatment.