Integrating Pretrained CNN and SVM for Improved Lung Cancer Classification

Lung Cancer is nowadays becoming a very common disease and a significant reason for death among Indians. Lung cancer often comes in two different forms: small- cell lung cancer and non-small-cell lung cancer. Among all other factors, Smoking cigarettes is one of the major reasons for lung cancer. The body is severely harmed by lung cancer, which is one of the most prevalent and dangerous cancers today. Therefore, the only method of treating lung cancer that is truly successful is early detection and diagnosis. Many lives would be saved if it should be diagnosed at an early stage. Lung Carcinoma, an uncontrolled malignant tumor characterized by uncontrolled cell growth in lung cells, is the alternative name for lung cancer. The paper aims to classify lung cancer in CT scan images. The greatest imaging technique available for medical use, a CT scan can be difficult for physicians to interpret and use to identify cancer. This paper proposes a technique that uses a hybrid method by combining the Support Vector Machine algorithm with CNN (Convolutional neural network) and gives better accuracy. CT scan images of patients are classified as to what type of lung cancer by using this technique. This technique efficiently classifies the cancer into its types. Using a hybrid model (Support Vector Machine algorithm with CNN technique) is an effective way to correctly identify the stages of lung cancer and pursue more accurate findings.