Enhancing Breast Cancer Detection: A Hybrid Approach Integrating Local Binary Pattern Features and Deep Learning Insights from Mammogram Images

Early identification of breast cancer improves treatment outcomes and lowers mortality rates. Mammogram images are useful for diagnosis, but their interpretation can be difficult and time-consuming. The current study analyzes the feasibility of promoting handmade and deep learning features to enhance the accuracy of breast cancer identification using mammography pictures. Previously, manual feature extraction has been labor-intensive and inconsistent. Furthermore, deep learning systems frequently suffer from limited data and architectural inefficiencies. To overcome these problems, we provide a novel strategy that makes use of both local binary pattern (LBP) features and automatic feature extraction from seven deep learning models. The concatenated LBP97.5%, and SVM and KNN classifiers trained on the hybrid feature beat existing state-of-the-art models. Our findings indicate the usefulness of this hybrid feature technique. This work demonstrates the potential of the suggested feature extraction strategy in improving classifier performance for breast cancer identification from mammography images. Our technique shows promise for early and more accurate diagnosis, contributing to better patient outcomes in the fight against breast cancer