A Machine Learning Framework for Lung Cancer Detection and Risk Assessment

Abstract

Cancer is a significant global health issue that impacts individuals across all newlinesocio-economic backgrounds. Numerous forms of cancer exist, including head and neck newlinecancer, brain cancer, stomach cancer, breast cancer, and pancreatic cancer. Lung cancer, newlinein particular, stands out due to its prevalence and high mortality rate, making it the most newlinecommon and deadliest form of cancer globally. Due to the increased air pollutions in newlinedeveloping countries nowadays, lung cancer has accelerated in the past decade. This newlinescary scene has necessitated the prevention and prompt identification of lung cancer. The newlineprimary approach for Detecting lung cancer in its initial stages involves Acquiring chest newlineComputed Tomography (CT) scans and scrutinizing them for anomalies. Skilled medical newlineprofessionals review these CT scans, which utilize X-ray technology, to identify any lung newlinenodules. However, as the incidence of cases rises annually, this examination process has newlinebecome increasingly burdensome, placing strain on healthcare systems. The development newlineof Computer-Aided Diagnosis (CAD) technologies has benefited clinicians treating lung newlinecancer. Lung nodule diagnosis and detection are aided by CAD systems. It is imperative newlineto detect lung nodules and classify whether the detected nodules are benign or malignant. newlineWhen there are thousands of images to be scrutinized, CAD systems come to the rescue newlineand are instrumental in preventing the deaths of millions of lung cancer patients since the newlinedisease can be detected and diagnosed accurately newline