eng Science and Education Publishing Journal of Computer Sciences and Applications 2328-725X 2026-05-20 14 1 21 30 10.12691/jcsa-14-1-4 JCSA20261414 article Dr. A.Shilpa Gupta a.shilpagupta@gmail.com 1 M. Uday Reddy 1 K. Jayanth Kumar Reddy 1 P. Medha Goud 1 T. Harshitha Reddy 1 Aditi Jopat 1 Department of Computer Science Engineering, Keshav Memorial College of Engineering, Ibrahimpatnam, Telangana, India We introduce MMedRAG++,a medical multimodal retrieval system that uses fusion-based representation learning and graph-based reranking (Graph-RAG) to improve on conventional Retrieval-Augmented Generation (RAG). Unlike baseline systems, which do not incorporate fusion strategies or graph-based reranking, MMedRAG++ improves cross-modal embeddings and retrieval coherence. Experiments are conducted primarily on PMC-OA, a challenging dataset with only ~10% unique captions and many unrelated image-text pairs, and IU-Xray is used for modality- specific subtasks. Graph-RAG demonstrates improved retrieval centrality and diversity, and fusion strategies, including Cross- Attention and DeepSet Fusion, enhance embedding quality. Quantitative evaluation on PMC-OA and IU-Xray confirms improved retrieval coherence and cross-modal alignment over baseline configurations. Top-1 Accuracy: 18.7%, Top-10 Accuracy: 57.4%. https://pubs.sciepub.com/jcsa/14/1/4/jcsa-14-1-4.pdf Medical AI RAG Graph-RAG Multimodal Fusion Contrastive Learning Medical Image-text Retrieval