An Intelligent Framework of Microcalcifications Breast Cancer Detection Using Heuristic Approach Aided Residual Attention Network

Background: Breast cancer detection has advanced in recent years, and one such recent technique is detecting the presence of Microcalcifications in the breast tissue from Mammogram images. Also, it is crucial to examine the favorable outcomes to examine whether the microcalcification samples to benign or tumor. Methods: A deep learning-based breast cancer microcalcifications detection model is developed in this work. The initial step involves in gathering images from reputable sources to ensure a diverse and comprehensive dataset. Following collection, these images are given to the detection phase, where the detection process is done using Adaptive Residual Attention Network (ARAN). During this phase, the parameters of the ARAN are optimally tuned using the Pine Cone Optimization Algorithm (PCOA) to ensure better performance. Result: Overall 92% of accuracy attained in the developed model. Conclusion: Experimental analysis is conducted on this deep learning-based model to validate its effectiveness in microcalcifications detection.