A comparison between different classification algorithms for predicting metastasis in breast cancer patients

Arman Behnam

Abstract: Breast cancer is one of the most common cancers among women around the world. According to World Health Organization (WHO), breast cancer is second reason for cancer mortality. Approximately 30%- 40% patients suffering from breast cancer will experience recurrence and 10%-15% of them were reported to die of cancer metastasis. Early diagnosis or prediction of metastasis will reduce mortality rate and treatment cost. In this study we have used a data set containing 555 record of patients with breast cancer (83 have experienced metastasis) and 8 features. Several machine Learning algorithms including Random Forest (RF), Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Support Vector Machine (SVM), K-Nearest Neighbors (KNN) and Multi-Layer Perceptron (MLP) were used to predict metastasis. Total accuracy, sensitivity, specificity, precision, recall, f1- score and area under curve (AUC) extracted out of Receiver operating characteristic values were used to evaluate models. The results show that Multi-Layer Perceptron Outperform other methods to predict the metastasis.