Electronic noses are bioinspired instruments that mimic the biological sense of smell. They are based on the use of gas sensors combined with pattern recognition methods. The analysis of patients’ breath odors has had a long history of application for the detection of various human diseases. Due to an increase in awareness that the early detection of diseases greatly increases the chances for successful treatment, there is an urge in demand for inexpensive, non-invasive, simple, and fast early qualitative diagnosis of diseases. The study is aimed to emphasize on the possibilities of combining an electronic nose device based on five tin oxide (SnO₂) sensors with a pattern recognition method to discriminate between healthy controls (HC), patients with liver cirrhosis (LCi), and gastric cancer (GCa). Breath samples were collected from 36 volunteers, including 13 HC, 15 LCi patients, and 8 GCa patients. For this purpose, pattern recognition techniques such as principal component analysis (PCA), discriminant function analysis (DFA), and support vector machines (SVM) are utilized for data processing of multivariable responses generated by the sensor array. The chemometrics results showed good discrimination between the data points of breath samples related to the three groups. The PCA score accounted 96.42% of the total variance, while scores of 100% were obtained using DFA and SVM for the recognition of the analyzed groups, respectively. This pilot study reveals that e-nose technology based on exhaled breath analysis could be an effective and promising non-invasive way to distinguish LCi and GCa patients from HC.