Predicting the recurrence of breast cancer using machine learning algorithms

Abstract

Breast cancer is one of the most common types of cancer among Jordanian women. Recently, healthcare organizations in Jordan have adopted electronic health records, which makes it feasible for researchers to access huge amounts of medical records. The goal of this study is to predict the recurrence of breast cancer using machine learning algorithms. We developed a Natural Language Processing algorithm to extract key features about breast cancer from medical records at King Abdullah University Hospital (KAUH) in Jordan. We integrated these features and built a medical dictionary for breast cancer. We applied multiple machine learning algorithms on the extracted information to predict the recurrence of breast cancer in patients. Our predicted results were approved by specialist physicians from KAUH. The medical dictionary was created and the accuracy of the data had been validated by targeted users (physicians, researchers). This dictionary can be used for personalized medicine. All machine learning algorithms had a nice performance. OneR algorithm has the best balance of sensitivity and specificity. The medical dictionary will help physicians to choose the most appropriate treatment plan in a short time. The machine learning prediction results can help physicians to make the correct clinical decision regarding their treatment options.