Exploring The Efficacy of Multi-modal LLM In Endometrial Cancer Diagnosis

Endometrial Carcinoma (EC) remains a significant diagnostic challenge due to its heterogeneous histopathology and clinical presentation. Recent advances in Visual Language Models (VLMs) have shown promise in medical imaging. One of the emergent technologies involves to explore large language models (LLMs) abilities with multimodal data. In this research, we find how Integration of VLMs in medical imaging analysis offers substantial benefits, includes better diagnostic accuracy, enhanced clinical decision support. This study demonstrates the efficacy of a multi-modal VLM framework for automated detection and classification of Endometrial Cancer stages, achieving 92% accuracy. Our approach integrates Magnetic resonance image (MRI) of endometrial Cancer and clinical text data to enhance diagnostic decision-making.