Breast DCE MRI Registration Using Golden Jackal Optimization Algorithm

Breast cancer is one of the most frequent cancers in women globally. Accurate breast magnetic resonance imaging (MRI) registration is crucial for early diagnosis, tumor development tracking, and assessing therapeutic success. This study examines the performance of three advanced meta-heuristic algorithms for rigid body registration of breast MRI images: Gorilla Troops Optimizer (GTO), Biogeography-Based Optimization with Elitism (BBO EL), and Golden Jackal Optimization (GJO). These methods are used to calculate the necessary transformation parameters for translation and rotation in order to achieve precise image alignment. The evaluation is based on a dataset containing multiple 2D breast MRI picture pairs. Normalized Mutual Information (NMI) is a common similarity metrics used to evaluate performance. Statistical significance is also established using the Wilcoxon signed-rank test methods. Experiments demonstrate that the GJO-based registration strategy outperforms GTO and BBO EL on multiple assessment measures. This study highlights the potential of swarm intelligence-based optimization methodologies for improving image registration in breast cancer diagnosis, contributing to the development of robust tools for oncology imaging.