SIGNATURE WORK
CONFERENCE & EXHIBITION 2024

As one of the most prevalent diseases affecting women globally, breast cancer requires effective diagnostic tools for early detection and management. Mammography has emerged as a critical imaging technique in this domain, offering a non-invasive method to detect breast abnormalities. However, the reliance on human interpretation can introduce errors, potentially leading to misdiagnoses or delayed treatments. To address these challenges, computer-aided diagnosis (CAD) systems have been developed to enhance the accuracy and sensitivity of mammogram analysis. While numerous large language models exist in the medical field, few incorporate a multi-modal approach that integrates professional analysis of breast cancer with CAD systems. This work introduces an innovative interactive system that combines a deep learning algorithm for mammogram detection, a specialized text-based language model for breast cancer, and a comprehensive multi-modal model. This integration aims to revolutionize breast cancer diagnostics, offering a more robust and nuanced analysis while minimizing human error and improving diagnostic outcomes.