2025/03/03 by Nurudeen, A. A., Umar, U. M., Asare, B.K. +1
#Breast cancer #Ensemble classifier #Machine learning #Performance #Recurrence #Survivability
paper · doi:10.60787/tnamp.v21.474
Worldwide, breast cancer is currently the most common cancer, accounting for 12.5% of all new annual cancer cases and it is one of the leading causes of cancer-related death in women second only to lung cancer. Incorporating machine learning (ML) classifiers into predicting the recurrence and survival of patients with breast cancer has emerged as a promising approach to enhance performance metrics. This study analyzed dataset on breast cancer obtained from clinical studies- Barau Dikko Teaching Hospital, Kaduna and when the performance of employed Conventional ML classifiers- Artificial Neural Network (ANN), Support Vector Machine (SVM), K-Nearest Neighbor (KNN) and Proposed ensemble learning classifier (ANN-KNN) was evaluated, it was observed that both Conventional ML and Proposed ensemble learning classifiers could predict recurrence of breast cancer and survivability of breast cancer patients. However, the performance of these conventional ML classifiers and the proposed ensemble learning classifier were compared. The results showed that the proposed ensemble learning classifier outperformed ML classifiers with 97% and 91.04% accuracy on recurrence and survival prediction of breast cancer respectively. It remains the best classifier in predicting breast cancer patients' recurrence and survivability, followed by the ANN classifier with accuracy of 90.5% and 81.93% respectively on recurrence and survivability prediction of breast cancer patients. The findings demonstrate that ensemble learning can enhance the performance of weak classifiers like SVM and KNN. Further extensive evaluation of other ML classifiers like decision tree and random forest can be performed using some combinations that can predict the recurrence and survivability of breast cancer patients with greater accuracy.