The goal of this proposal is to build a computer algorithm to aid making reliable prognostics in early stage lung cancer. Lung cancer is typically detected at late stages ad TNM stage III or IV often associated with poor prognosis. Early stage lung cancer is however present as small lesion (a coin size lesion) named stage T1 lesion. It can be present as an isolated nodule which is not aggressive (TN1N0) or present with nodal metastases (T1N1) which is often associated with aggressive malignancy and therefore poor prognosis. But each type can be present with varying size, shape and molecular signatures that can dictate their behavior and aggressiveness. Currently for early stage lung cancer, clinicians make prognostic decision based on the pathologic staging which sometimes is personal and not accurate. The objective of this study is to develop a computerized algorithm that can integrate pathologic, environmental (tissue environment) and molecular markers to improve prognosis accuracy and decide on best treatment and care. Three aims are proposed: Aim 1: Improve the clustering based algorithm and link with current database and validate the results by testing it. Aim 2: Modify the programing for implementation on personal computer and IPads to make to facilitate use by clinicians. Aim 3: Create a logistic Bayesian regression to better assess and characterize early stage lung cancer and establish contribution of newly identified prognostic markers.