Enhancing Lung Cancer Diagnosis with Machine Learning Methods and Systematic Review Synthesis

Lung cancer is the leading cause of death in the USA It is caused for various reasons, such as malignant tumors that grow spontaneously in the lung and spread to other parts of the body, where they can be devastating. Other than that, there are so many external reasons why smoking is the primary factor that gradually creates lung problems that lead to lung cancer. However, we can prevent this deadly cancer by detecting it at an early stage. A machine learning-based detection model has the potential to detect lung cancer, which helps to predict the early stage of cancer. So, we aim to build some powerful machine learning algorithms to predict this cancer that can help physicians make decisions about the diagnosis process of a particular patient, whether the patient needs a high or a regular diagnosis level. This can reduce the treatment cost of a patient because the physician knows from the result how many doses are needed for a patient, so it can avoid unnecessary treatment, which saves money and brings an efficient diagnosis system. We aimed to build a sustainable model that could predict the disease accurately, and finally, our models, such as XGBoost, achieved the highest value of other models. According to the accuracy level, XGBoost performed up to 95.92%, LightGBM, AdaBoost, Logistic Regression, and Support vector machine achieved 91.50%, 93.32%, 57.41 %, and 89.32% accuracy, respectively.