An efficient pulmonary carcinoma nodule detection model

Lung cancer is among the top causes of death globally, significantly impacting global health due to its high incidence and mortality rates. Breathing difficulties and reduced quality of life are common in people with respiratory illnesses such as asthma, interstitial lung disease, and chronic obstructive pulmonary disease (COPD). Doctors diagnose and establish the stages of cancer, which may not always be correct. The only way to increase the chances of human survival is to recognize them early. The average survival rate increased from 14% to 49% when lung cancer was diagnosed early. Although computed tomography (CT) is significantly more effective than radiography, a complete diagnosis requires a combination of imaging methods. Machine learning technology has been developed and tested for lung cancer detection using CT images. Image processing and machine learning techniques for lung cancer identification were used to a dataset of CT scans in order to categorize the presence of lung cancer. From kaggle, we obtained a lungs imaging dataset. Fuzzy C-means, its variants and K-means algorithms were considered for image segmentation. Aberrant images were segmented to concentrate on the tumor. SVM, KNN, RF, and CNN were employed to study the classification efficiency and to determine if the CT image of the patient is normal or abnormal. The study reveals that the EnFCM method of Segmentation followed by CNN showed 99% accuracy, EnFCM segmentation followed by classification in tumor detection and identification has a good impact. The accuracy obtained with this model is more efficient.