The investigators investigate the utility of FDG PET/CT based radiomics in lung cancer, including diagnosis and prognosis.

Recent studies have shown that, in addition to inter-tumor heterogeneity, tumors often display startling intratumoral heterogeneity in various features including histology, gene expression, genotype, and metastatic and proliferative potential, which is often associated with adverse tumor biology. Unfortunately, it is difficult to assess intratumoral heterogeneity with random sampling or biopsy as this does not represent the full extent of phenotypic or genetic variation within a tumor. Given the limitations of current biopsy strategies, there is an important potential for medical imaging, which has the ability to capture intratumoral heterogeneity in a non-invasive way.

Borrowed from the concept in genomics and/or proteomics, radiomics was specifically proposed for medical or radiological images. It is a promising technique for improving diagnosis, staging, prognosis, treatment response prediction and potentially allowing personalization of cancer treatment. It is a process of extraction and analysis of high-dimensional image features from radiological images obtained with CT, MR or PET, which could be either qualitative or quantitative. The basic assumption of radiomics is that tumor biology could be captured by radiomic features .

The purpose of this study is to investigate the utility of FDG PET/CT based radiomics in lung cancer. Four PET/CT centers will be involved in this study, in which more than 1000 patients diagnosed as lung cancer will be retrospectively enrolled.

Inclusion Criteria:

- All patients diagnosed as lung cancer patients who had a FDG PET/CT scan before treatment between 1 Jan, 2013 and 30 December, 2016 in the four collaborative hospitals.

Exclusion Criteria:

- The patient without follow-up information

All patients diagnosed as lung cancer patients who had a FDG PET/CT scan before treatment between 1 Jan, 2013 and 30 December, 2016 in the four collaborative hospitals.

there is no plan to make individual participant data (IPD) available to other researchers.