The causes of the lung cancer to be common cause of the death among the people throughout the world. This is one of the diseases that causes the deadliest impacts which can also cause fluid to accumulate around the lungs, making it harder for the affected lung to expand fully when some-one inhale. This work has introduced one of the best ways to detect the lung cancer and the methods that are used to detection to increase the accuracy and yield and decrease the diagnosis time. The major things in this work consider two datasets, including the Lung Image Database Consortium and Image Database Resource Initiative. In this work it basically considers the two dataset the main dataset and public dataset. The identification of lung cancer a novel diagnosis method based on the Deep Transfer Convolutional Neural Network (DTCNN) and Extreme Learning Machine and Extreme is explored which has been trained with the Image Net dataset beforehand. When this work compares between DTCNN and ELM is important for both clinical care and secondary analysis. Although multiple applications have been developed for computational phenotyping in lung cancer, distant recurrence identification still relies heavily on manual chart review. Two datasets, including the Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) public dataset for LIDC-IDRI dataset, the experimental results Shows different type of results but the performance of the novel DTCNN-ELM model achieved the performance with an accuracy, a sensitivity, a specificity, an area under the receiver operator curve (AUC) and testing time per which has the most reliable results compared with current state-of-the-art methods.