Efficient segmentation and classification of the lung carcinoma by deep learning approches

Abstract

Lung carcinoma is a cancerous disease that forms in lung tissue. Since lung cancer tends to spread or metastasize very early in its course, it is a very life-threatening cancer and one of the most difficult cancers to treat. According to the World Health Organization (WHO), lung cancer has been the most common cancer in the world for several decades. Detecting and treating lung cancer at an early stage is very important. This can lead to more treatment options, less invasive surgery, and a higher survival rate. Cancer usually develops in the lungs either by developing in the lungs or by spreading to the lungs. Cases that start in the lungs are categorized as primary lung cancer, and cases that spread to the lungs from another part of the body are categorized as secondary lung cancer (the metastasis of cancer to the lungs). In general, lung cancer mainly refers to primary lung cancer. Primary lung cancer is divided into two main types namely, Non-Small Cell Lung Cancer (NSCLC), and Small Cell Lung Cancer (SCLC). In current medical diagnosis, treatment, and surgery, medical imaging plays one of the most important roles, since imaging devices such as Computed Tomography (CT), Magnetic Resonance Imaging (MRI), and ultrasound diagnostics with the technical support of CAD systems also yield a great deal of information about diseases and organs. In the first stage of this work, an Efficient Segmentation and Classification of the LUNG carcinoma using edge-based segmentation and CNN for the malignant tumor identification is introduced. Here, the lung carcinoma is identified by other effective methods like an adaptive median filter for preprocessing, Histogram equalization, edge-based segmentation, and Convolutional Neural Network (CNN) for effective classification. The Adaptive Median Filter and Histogram Equalization method are employed to enhance segmentation efficiency. In this process, Edge-based segmentation and CNN for classification techniques get introduced for lung carcinoma detection. The edge-segmentation method grows for efficient image segmentation in regions. Further, the essential characteristics can be derived from the similarity of the features and then transferred to the classification method. newline