Optimizing Lung Cancer Diagnosis with ML Classification Methods

Abstract

One of the most advanced causes of deaths due to cancer that occurs all over the world is lung cancer since such cancer is not easily diagnosed in early stages and it takes advance staging so that the disease can be diagnosed. This research study is proposed to learn how machine learning system assists the artificial intelligence to foresee the invasion of lung cancer by early staging through supporting the symptoms and the lifestyle data that have been reported by the patients. A well-organized record of clinical and behavioral features was a preprocessed dataset to ensure that the model was ready through feature encoding and normalization. To categorize the incidence of lung cancer, they have employed some of the supervised learning algorithms such as logistic regression (LR), decision trees (DT) and ensemble strategies. AdaBoost model showed better results than the others and was therefore used in classification of the three types of lung cancer, i.e. the three namely, the small cell carcinoma, adenocarcinoma and the large cell carcinoma. As depicted in the comparison research, it was uncovered how ML could be utilized as a beneficial diagnostic procedure to ascertain the risk of lung cancer and thus be capable of prompt and more informed medical treatment. This study reflects the significance of data-driven practices in completing the existing diagnostic systems and aiding the clinical decision-making in oncology.