A Comparative analysis study of lung cancer detection and relapse prediction using XGBoost classifier
Rana Dhia’a Abdualjabar, Osama A. Awad
Abstract
Abstract Lung cancer is the leading cancer for causing death for both men and women. It also has one of the lowest survival rates in five-year of all cancer types. It remains a challenge to lung cancer relapse prediction after surgery, especially for non-small cell lung cancer (NSCLC). This study aimed to enhance prediction and detection using eXtreme Gradient Boosting (XGBoost) model to detect lung cancer diagnoses and predict its relapse after surgery by using gene expression and its transcriptome changes due to cancer. This can aid to enhance early tumour progression handling and reducing the painful treatment. In this study, it used real New Generation RNA_seq (NGS) and microarray gene expression datasets for different types of lung cancer. The results demonstrated the effectiveness of the XGBoost model compared to other machine learning models especially in handling unbalance datasets.