Lung Cancer Patient’s Survival Prediction Using GRNN-CP

Published results for cancer patients have been previously estimated by applying various machine learning techniques to large. especially, for lung cancer, it is not well known to the time, which sorts of techniques would generate more imminent information, and which data attributes should be employed in order to prepare this information. In this study, a supervised learning technique is implemented to analyze lung cancer patients in terms of survival, the purpose of this study is to predict lung cancer and to compose an aiding model that will help form a more reliable prediction as a factor that is vital for advancing survival time evaluation. We utilize general regression neural networks (GRNN) for replacing the regular predictions with prediction periods to achieve a moderate percentage of confidence. The mechanism applied here employs a machine learning system called conformal prediction (CP), to assign consistent confidence measures to predictions, which are combined with GRNN. We apply the resulting algorithm to the problem of lung cancer diagnosis of supervised learning techniques is applied to the NCI database to classify lung cancer patients. Experimental results confirm that the prediction formed by this method is feasible and could be useful in clinical institutions.