Graph-Enhanced Medical Question-Answering System Integrating Knowledge Graphs and Large Language Models
DOI:
https://doi.org/10.55578/joaims.260408.001Keywords:
Knowledge graph (KG), Graph algorithms, Large language model (LLM), Intelligent question answeringAbstract
Objective: To address the challenges posed by the rapid growth of medical data and the fragmentation of knowledge, this study aims to construct a medical knowledge graph (KG) and provide efficient knowledge services for clinical practice.
Methods: A total of 44,157 entities and 291,170 relationships from an open-source database were integrated to build a local medical KG based on Neo4j. Graph algorithms including degree centrality, Louvain community detection, K-nearest neighbor, and Dijkstra’s algorithm were applied to analyze the data. The retrieval results from the KG were combined with the Spark Lite model from IFlytek to develop a dual-channel question-and-answer system.
Results: Highly related entities, such as acute urethritis and blood routine tests, were successfully identified. The analysis yielded 35 disease communities and 17 department communities. Highly similar disease pairs, such as “lung abscess” and “pulmonary bullae,” were discovered. Potential therapeutic pathways, such as “Erythromycin Ethylsuccinate Granules - Erythrasma,” were uncovered, revealing clinical associations among various entities. The system is accessible at http://yangbiolab.cn:8054/.
Conclusion: Graph algorithms effectively mine key patterns and potential associations within medical knowledge, with several findings aligning closely with clinical practice. The integrated system offers an intuitive platform for exploring medical knowledge.
References
[1] L. Wang, H. Hao, X. Yan, T.H. Zhou, K.H. Ryu, From biomedical knowledge graph construction to semantic querying: a comprehensive approach, Sci. Rep. 15 (2025), 8523.
[2] G. Zhao, W. Gu, W. Cai, Z. Zhao, X. Zhang, J. Liu, MLEE: a method for extracting object-level medical knowledge graph entities from Chinese clinical records, Front. Genet. 13 (2022), 900242.
[3] H.-C. Hsueh, S.-C. Chien, C.-W. Huang, H.-C. Yang, U. Iqbal, L.-F. Lin, et al., A novel multi-level refined (MLR) knowledge graph design and chatbot system for healthcare applications, PLoS ONE. 19 (2024), e0302620.
[4] H. Zhu, A graph neural network-enhanced knowledge graph framework for intelligent analysis of policing cases, Math. Biosci. Eng. 20 (2023), 11585–11604.
[5] X. Mo, J. Pang, B. Wan, R. Tang, H. Liu, S. Jiang, Multirelational graph contrastive learning with learnable graph augmentation, Neural Netw. 181 (2025), 106757.
[6] R. Johnson, M.M. Li, A. Noori, O. Queen, M. Zitnik, Graph artificial intelligence in medicine, Annu. Rev. Biomed. Data Sci. 7 (2024), 345–368.
[7] X. Zhou, S. Zhang, M. Agarwal, J. Akryod, S. Mosbach, M. Kraft, Marie and BERT—a knowledge graph embedding based question answering system for chemistry, ACS Omega. 8 (2023), 33039–33057.
[8] H. Park, J. Son, J. Min, J. Choi, Selective UMLS knowledge infusion for biomedical question answering, Sci. Rep. 13 (2023), 14214.
[9] C. Fu, X. Pan, J. Wu, J. Cai, Z. Huang, F.v. Harmelen, et al., KG4NH: a comprehensive knowledge graph for question answering in dietary nutrition and human health, IEEE J. Biomed. Health Inform. 29 (2025), 1793–1804.
[10] Y. Feng, L. Zhou, C. Ma, Y. Zheng, R. He, Y. Li, Knowledge graph-based thought: a knowledge graph-enhanced LLM framework for pan-cancer question answering, GigaScience. 14 (2025), giae082.
[11] A. Singhal, Introducing the knowledge graph: things, not strings, Official Google Blog, 2012.
[12] Q. Wang, Z. Mao, B. Wang, L. Guo, Knowledge graph embedding: a survey of approaches and applications, IEEE Trans. Knowl. Data Eng. 29 (2017), 2724–2743.
[13] M. Afshar, Y. Gao, D. Gupta, E. Croxford, D. Demner-Fushman, On the role of the UMLS in supporting diagnosis generation proposedby Large Language Models, J. Biomed. Inform. 157 (2024), 104707.
[14] K. Schatz, D. Korn, A. Tropsha, R. Chirkova, Workflow for domain- and task-sensitive curation of knowledge graphs, with use case of DRKG, 2022 IEEE International Conference on Big Data (Big Data), IEEE, Osaka, Japan, 2022, pp. 3692–3701.
[15] P. Chandak, K. Huang, M. Zitnik, Building a knowledge graph to enable precision medicine, Sci. Data. 10 (2023), 67.
[16] H. Wang, Q. Zu, M. Lu, R. Chen, Z. Yang, Y. Gao, et al., Application of medical knowledge graphs in cardiology and cardiovascular medicine: a brief literature review, Adv. Ther. 39 (2022), 4052–4060.
[17] Y. Lan, S. He, K. Liu, X. Zeng, S. Liu, J. Zhao, Path-based knowledge reasoning with textual semantic information for medical knowledge graph completion, BMC Med. Inform. Decis. Mak. 21 (2021), 335.
[18] D.Y. Li, T.J. Hu, J.L. Li, Q. Qian, W.Y. Zhu, Construction and application of the Chinese unified medical language system, J. Intell. 30 (2011), 147–151.
[19] H.Y. Zan, H.Y. Dou, Y.X. Jia, T.F. Guan, B. Odbayar, K.L. Zhang, et al., Construction of Chinese medical knowledge graph based on multi-source texts, J. Zhengzhou Univ. (Nat. Sci. Ed.), 52 (2020), 45–51.
[20] H. Long, Y. Zhu, L. Jia, B. Gao, J. Liu, L. Liu, et al., An ontological framework for the formalization, organization and usage of TCM-knowledge, BMC Med. Inform. Decis. Mak. 19 (2019), 53.
[21] X. Xue, Y. Mei, B. Zhao, M. Zhang, Adaptive similarity feature construction for ontology matching via multilayer hybrid genetic programming, IEEE Trans. Evol. Comput. 30 (2026), 519–533.
[22] T. Cheng, T. Bi, W. Ji, C. Tian, Graph convolutional network for image restoration: a survey, Mathematics. 12 (2024), 2020.
[23] R. Zeng, C. Zeng, X. Wang, B. Li, X. Chu, Incentive mechanisms in federated learning and a game-theoretical approach, IEEE Netw. 36 (2022), 229–235.
[24] L. Ma, M. Huang, S. Yang, R. Wang, X. Wang, An adaptive localized decision variable analysis approach to large-scale multiobjective and many-objective optimization, IEEE Trans. Cybern. 52 (2022), 6684–6696.
[25] T. Winograd, Five lectures on artificial intelligence, Stanford University, Stanford, CA, USA, 1974.
[26] J. Kupiec, MURAX: a robust linguistic approach for question answering using an on-line encyclopedia, Proceedings of the 16th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, ACM, New York, NY, United States, 1993, pp. 181–190.
[27] L. Dong, M. Lapata, Language to logical form with neural attention, Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, ACL, Berlin, Germany, 2016, pp. 33–43.
[28] R. Jia, P. Liang, Data recombination for neural semantic parsing, Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, ACL, Berlin, Germany, 2016, pp. 12–22.
[29] R. Wang, M. Wang, J. Liu, W. Chen, M. Cochez, S. Decker, Leveraging knowledge graph embeddings for natural language question answering, In: G. Li, J. Yang, J. Gama, J. Natwichai, Y. Tong (Eds.), Database Systems for Advanced Applications, DASFAA 2019, Lecture Notes in Computer Science, Vol. 11446, Springer, Cham, 2019, pp. 659–675.
[30] M.Y. Cao, Q.Q. Li, Z.H. Yang, L. Wang, Y. Zhang, H.F. Lin, et al., A question answering system for primary liver cancer based on knowledge graph, J. Chin. Inf. Process. 33 (2019), 88–93.
[31] Y.Y. Chen, Construction and research of knowledge graph for stroke, syndrome, disease and medicine based on deep learning, Master’s Thesis, Gansu University of Chinese Medicine, Lanzhou, China, 2023.
[32] Z.M. Yi, K. Li, Z.Y. Xu, Z.M. Xu, F. Gao, A construction method and device for personalized medication assistance question answering system for Parkinson’s disease, Patent CN202410698116.0, Beijing, China, 2024.
[33] L.C. Freeman, Centrality in social networks conceptual clarification, Soc. Netw. 1 (1978–1979), 215–239.
[34] V. Blondel, J.-L. Guillaume, R. Lambiotte, Fast unfolding of communities in large networks: 15 years later, J. Stat. Mech. Theory Exp. 2024 (2024), 10R001.
[35] S. Lloyd, Least squares quantization in PCM, IEEE Trans. Inf. Theory. 28 (1982), 129–137.
[36] E.W. Dijkstra, A note on two problems in connexion with graphs, Numer. Math. 1 (1959), 269–271.
[37] H. Lu, M. Halappanavar, A. Kalyanaraman, Parallel heuristics for scalable community detection, Parallel Comput. 47 (2015), 19–37.
[38] D. Liben-Nowell, J. Kleinberg, The link-prediction problem for social networks, J. Am. Soc. Inf. Sci. Technol. 58 (2007), 1019–1031.
[39] B. Perozzi, R. Al-Rfou, S. Skiena, DeepWalk: online learning of social representations, Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, ACM, New York, NY, United States, 2014, pp. 701–710.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Yiqin Huang, Qing Liu, Zhengru Xie, Baopeng Ding, Ziwei Wang, Fenglong Yang (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
