Machine Learning Prediction of Cancer from The Publicly Available Dataset

Abstract

Prostate cancer in the second most common cause of cancer in men around the world and in Qatar with a high incidence rate worldwide. This has resulted in an increased mortality rate, making prostate cancer a healthcare burden. Early detection of prostate cancer is crucial in reducing mortality; however, the current detection procedures are invasive with prostate cancer screening test not being easily accessible. This has led to the development of machine learning approaches in detection of cancer with aims to improve healthcare accuracy and patient outcomes. This study examines the efficacy of machine learning model in prediction of prostate cancer using publicly available healthcare dataset. It aims to determine the best classifier algorithm and to develop a standard operating procedure (SOP) that can be used in a machine learning model for prostate cancer prediction. Lastly, this study examines the main feature class based on machine learning model that can increase the risk of developing prostate cancer.