CIS Research Centre is proud to support Dr. Abubakar Siddique by providing funds for a research assistant for his project Towards Accurate and Explainable Breast Cancer Diagnoses using Machine Learning.
Breast cancer remains one of the most prevalent and life-threatening malignancies worldwide, including in New Zealand. Although mammography is a cost-effective screening method, the complex patterns in mammograms make accurate diagnosis challenging. Deep learning models have shown promise but often suffer from limited interpretability and robustness. In contrast, rule-based machine learning models enable interpretable decision-making by linking predictions to specific mammogram features, providing a promising approach to enhance diagnostic transparency, reliability, and clinical trust.
This work aims to develop a novel rule-based machine learning system for breast cancer screening and diagnosis, which integrates code fragments derived from genetic programming into the extended supervised tracking and classifying system. By improving the flexibility of rule expression, this approach will significantly enhance the ability of the model to express complex nonlinear feature interactions. Experiments will be conducted on well-known mammography datasets (DDSM and MIAS), and results will be compared with the state-of-the-art systems. The decision-making process of the proposed system is inherently interpretable, which is a step towards explainable AI.