Identification and classification of lung nodule based on meshfree approach and deep learning model

Abstract

Lung cancer is a leading cause of mortality for both men and women, with pulmonary nodules being a key early detection sign. Lung cancer is a lethal disease caused by genetic disorders and metabolic abnormalities and is one of the leading causes of mortality globally. Lung and colon cancer are the most contributing factors to mortality and disability in patients. The diagnosis of lung cancer by histopathology is crucial for patient treatment. The present research encompasses the use of histopathological and CT scan pictures to identify cancerous tissues in the colon and lung. The current study addressed the improvement of CT scan images using a mesh-free approach and the categorization of colon and lung cancer using various models developed using convolutional neural networks. This thesis has investigated the categorization of lung and colon cancer tissues newlineusing a combination of pre-trained models and other conventional classification techniques. The proposed categorization approach has been validated using two different kinds of CT scan pictures and histological images from the colon and lung.