Previous studies have identified a large inter-observer variability in the definition of Gross Tumor Volume (GTV) in lung cancer due to for instance difficulties in discriminating between tumor and aelectasia, and influence of window settings on CT scan. Accurate GTV definition is a cornerstone in quality assurance in radiotherapy and any uncertainty will not only jeopardize the clinical outcome but also comparison between clinicians, institutes and more importantly different treatment modalities and follow-up. This study aims at developping an objective segmentation tool for automated contouring of lung lesions. A second goal is to use this tool as a quantative measure of perfusion, diffusion and metabolic activity, whicj could be indicative of the treatment outcome. A feasability study will be performes on pulmonary metastasis and primary lung tumors, and based on these results a second phase will be designed to investigate the applicability in hepatic metastasis. All patients treated with stereotactic body radiotherapy included in this study will receive PET scan as an objective bench-mark of the 'CT-approach". As such the multidisciplinary treatment approach in the management of cancer.