Lung cancer is the leading cause of cancer-related death in New Zealand, mainly because of the high cost of current lung-cancer screening methods, which results in late detection. Several studies have demonstrated scent-detection dogs’ ability to accurately identify cancer, but this technology has not been validated in clinical settings. I have developed and tested an automated scent-detection apparatus that will be used to train and evaluate scent-detection dogs for operational lung cancer screening using breath and saliva samples. Following a blind test designed to establish the diagnostic accuracy of the dogs, they will be evaluated at suitable points in the diagnostic pathway to determine what additional value they might offer. Improved economic feasibility of early screening programmes or improvements in accuracy or speed of diagnosis in existing diagnostic pathways would result in increased detection rates and reduced mortality for individuals affected by lung cancer.