Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{188061,
        author = {Manushri Bhuyan},
        title = {Knowledge-Enhanced MRI Diagnostic Assistant using Retrieval-Augmented Generation},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {7},
        pages = {730-735},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=188061},
        abstract = {The usage of Large Language Models in clinical scenarios has inherent difficulties, especially in sensitive domains like MRI interpretation, where the chances of factual inconsistency or hallucination are extremely high. RAG handles this limitation effectively by dynamically anchoring the generative output in the LLM through evidence-based, domain-specific medical knowledge. In the process, verified outside information has been utilized to transform the LLM from a general predictor into an evidence-based reasoning engine. A state-of-the-art example is the Knowledge-Enhanced MRI Diagnostic Assistant by means of Retrieval-Augmented Generation Platform, in which quantitative radiomic features of multi-sequence MRI modalities (such as T1 and T2-FLAIR) are combined with an LLM referencing a carefully curated and regularly updated medical knowledge base. The retrieval mechanism allows the model to generate diagnostic reports that are transparent, verifiable, and compatible with clinical standards. For this reason, this approach provides not only improved diagnostic accuracy but also higher consistency and auditability of radiological reports. Despite such benefits, challenges exist on how to optimize system latency and continuously index diverse medical literature. Future efforts will be channeled into advanced multi-modal retrieval strategies that incorporate imaging and textual features, and rigorous large-scale multi-institutional validation to cement RAG as a reliable and indispensable tool for AI-powered support in medical imaging and clinical decision-making.},
        keywords = {Retrieval-Augmented Generation, RAG, MRI Analysis, Medical Imaging, Large Language Models, Hallucination Mitigation, Explainable AI, Multimodal Reasoning},
        month = {December},
        }