AI General Physician Agent: A Multi-Base Intelligent System for Healthcare Assistant
Keywords:
Artificial intelligence, Conversational AI, Digital health, Healthcare assistance, Intelligent systems, Multi-ggent system, Natural language processingAbstract
Recent advancements in artificial intelligence have significantly improved the ability of systems to understand and respond to human language. In the healthcare domain, conversational assistants are increasingly used to provide quick guidance and improve access to basic medical information. However, many existing solutions rely on a single processing unit, which often leads to generalized responses and limited understanding of user needs across different health-related areas. This paper presents an intelligent multi-agent system designed to deliver personalized support for health and wellness. The proposed system divides responsibilities among specialized agents that focus on medical assistance, fitness guidance, and mental well-being. A central decision mechanism analyzes user input and directs each query to the most suitable agent based on its context. The system is developed using Python and integrates modern frameworks for managing agent workflows and natural language processing. It also includes a memory component that captures user information, such as preferences and health goals, to improve response relevance over time. Experimental evaluation indicates that the multi-agent approach enhances accuracy, adaptability, and user-specific recommendations compared to traditional single-agent systems. The study demonstrates the potential of intelligent agent-based systems in building accessible and reliable digital healthcare assistants. The AI General Physician Agent is a Multi-Base Intelligent System for healthcare assistants. It is an innovative collaborative framework that aims to improve diagnostic precision, operational efficiency, and tailored patient care. Instead of one chatbot, this solution uses a Multi-Agent System (MAS) concept of a team of specialist AI agents working together to handle patient journeys from triage to follow-up. Less Physician Burnout. Automated repetitive administrative activities like paperwork, coding, and prior authorization save physicians significant time. Improved Accuracy in a Multi-agent collaboration boosts diagnostic accuracy by eliminating errors related to a lack of information. Agents’ Personalized Care Planning examines genetic profiles, medical histories, and therapy reactions to prescribe targeted medicines.
References
T. Brown, B. Mann, N. Ryder, et al., “Language models are few-shot learners,” in Advances in Neural, Information Processing Systems, vol. 33, pp. 1877–1901, 2020.
J. Devlin, M. Chang, K. Lee, and K. Toutanova, “BERT pre-training of deep bidirectional transformers for language understanding,” in Proceedings of NAACL HLT, pp. 4171–4186, 2019.
H. Touvron, L. Martin, K. Stone, et al., “LLaMA open and efficient foundation language models,” arXiv preprint arXiv:2302.13971, 2023.
S. Russell and P. Norvig, Artificial Intelligence: A Modern Approach, 3rd ed., Pearson, 2010.
Y. Bengio, A. Courville, and P. Vincent, “Representation learning: A review and new perspectives,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 35, no. 8, pp. 1798–1828, 2013.
J. Wei, X. Wang, D. Schuurmans, et al., “Chain of thought prompting elicits reasoning in large language models,” in Advances in Neural Information Processing Systems, 2022.
Vasilios Mavroudis, “LangChain, vol.3,” 2024.
M. A. Rajput, “AI Frameworks and Platform Integration Services | Kodexo Labs,” Kodexo Labs
W. Qi, J. Pan, H. Lyu, and J. Luo, “Excitements and Concerns in the post-ChatGPT Era: Deciphering Public Perception of AI Through Social Media Analysis,” Telematics and Informatics, vol. 92, pp. 102158, Aug. 2024.
T. Simbini et al., “Digital Health Interventions in Strengthening Primary Healthcare Systems in Sub-Saharan Africa: Insights From Ethiopia, Ghana, and Zimbabwe,” PLOS Digital Health, vol. 5, no. 1, pp. e0000863, Jan. 2026.
