The incidence of lung cancer continues to increase. From patients diagnosed today with lung cancer, only 14% will survive five years later. However, if lung cancer is detected early, the five-year survival can rise from 1% for stage IV to 67% for stage IA. For this reason, several methods are investigated for an early detection of the disease, to thereby increase the chances of survival. The analytical and statistical approaches described here clearly provide a powerful means of selecting diagnostically significant compounds in expired breath. Applying these procedures to a larger sample population should help confirm, refute, or expand the significance of the compounds identified in this preliminary study and establish diagnostic criteria for general use. The classification function finally developed could then be used in a dedicated analytical system to classify unknown samples rapidly and reliably. The results suggest, however, that the measurement of a discrete number of compounds may be sufficient for identifying persons with lung inflammation.