Abstract
Pneumonia is the main reason which causes patients death in hospital. The novelty breath gas analysis method probability offers doctors to identify and monitor this clinical condition immediately. But the gas analysis equipments, such as gas chromatography (GC) and Fourier transform infrared spectroscopy (FTIR), are very accurate but expensive, time consuming, and nonportable. The electronic nose (e-Nose) which has advantages of low cost and easy operation is more suitable to realize an infection pneumonia early alert system for personal use. In this work, we collected the sampling gas from patient who got infection pneumonia and then had been healed respectively. We used the commercial e-nose to get the raw data, and analyzed by principal component analysis (PCA). The results show that we could distinguish between patients with infection pneumonia and patients who had been healed.