International Journal of Contemporary Research in Multidisciplinary

Abstract:

Lung cancer is a common cause of death for people all over the world. The most prevalent type of lung cancer has a significant fatality rate. Lung cancer screening can help identify cancer at an early stage if the sickness is recognized and treated while it is still in its early stages. As a result, early detection of lung cancer increases patients' chances of survival. Computed tomography (CT) scans of the lungs may be more accurate than X-rays and MRIs in detecting the presence of cancer. The lungs are scanned, and images are obtained using DICOM (Digital Imaging and Communications in Medicine). We apply a range of preprocessing techniques to enhance and smooth images. In this study, the region of interest for the lung tumor is segmented using edge detection and thresholding (ROI). Finally, we compute several geometrical features of the retrieved ROI using Support Vector Machine and classify them into benign and malignant severity levels (SVM). supervised machine learning methods called SVM classifiers are used to categorize data. They have the advantage of handling tiny samples of high dimensional data. Performance of the SVM is assessed using each feature as an input. As a result, a system that employs image processing techniques is used to diagnose lung cancer in CT scans.

Keywords:

Lung Cancer Detection, Lung Malignant Growth, Lung Diseases

Aim/Objectives

Methods