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{196969,
        author = {Tejas Ajayrao Padole and Vedant V. Girhe and Maifuz Sayyad and Ram Gawande and Dr. Sunil R. Gupta},
        title = {HealthAgent An AI-based Healthcare Assistant with RAG and Large Language Models},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {6380-6386},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=196969},
        abstract = {The rapid pace of Large Language Models (LLM) is capable of transforming the healthcare. access model; however, it is common that the conventional conversational AI is marked by delusions of reality and the lack of complex thinking in sensitive health situations. This paper is presenting Health agent which is a state-of-the-art healthcare assistant that will transform. its more traditional chatbot architecture to a more involved Agentic Retrieval- Augmented Generation. (RAG) framework. The proposed architecture has Llama 3.3 70B as the reasoning machine, and with the assistance of LangChain and LangGraph to facilitate stateful and multi-turn dialogue and autonomous task separation. HealthAgent, in order to ensure clinical grounding, introduces ChromaDB is a high-performance vector knowledge base that enables the system to access. validated medical data in making an inference and not merely to apply stochastic text generation. Such system has an agentic nature that allows it to automatically schedule parallel activities e.g., match clusters of symptoms with contraindications of medications and first-aid measures. with preratification. A high throughput FastApi back end is used as the technical architecture. to offer a React.js front end and low latency processing to offer a smooth experience with the patient. Experimental findings confirm the fact that Agentic RAG approach is significantly more effective. Thank the non-emergency baseline LLM models regarding the factual accuracy and misinformation. decrease.
The high-quality data privacy system and the robust data privacy framework are the most significant features of the System reliability. constant optimization. HealthAgent is concerned with the saftey of confidential patient data. The implementing the encryption protocols of the industry level and adhering to the helathcare requirements data security to offer protection of such information in all levels and stages of intercation. Besides, feedback loop integration helps the AI to learn in the circumstances of human users, enhancing its applicability in the discussion. Such security priorities and constant improvements bring about a level of trust, which is a component of mass use of AI. Supportive staff to the division of the health.},
        keywords = {Agentic RAG, Llama 3.3 70B, LangGraph, LangChain, ChromaDB, Vector Database, Large Language Models.},
        month = {April},
        }