Unsupervised Graph Learning on Live Health Data
DOI:
https://doi.org/10.5281/4axsb282Keywords:
Graph Neural Networks, FHIR, interoperable healthcare, unsupervised learning, real-time streams, anomaly detection, health state inference, data privacy.Abstract
Interoperability remains a critical challenge in healthcare, hindering data exchange across institutions and complicating the construction of comprehensive patient histories. Unsupervised Graph Neural Network (GNN) learning on streaming Fast Healthcare Interoperability Resources (FHIR) data has the potential to facilitate cross-institution inference, yet this approach remains unexplored. We present a framework capable of detecting anomalies in real-time FHIR data streams and inferring patients’ underlying health states, learning feature representations over time without explicit supervision. Particular attention is given to the privacy implications of the unsupervised approach and to GNN architecture selection. Promising results on both fronts indicate that the learned representations meaningfully capture disease-related information, opening new avenues for interpretability and knowledge transfer.
Broader deployment will require adapting to model drift, dynamic changes in population structure, and the absence of corresponding patient labels. This work represents a critical step toward unsupervised GNN applications in dynamic healthcare environments, demonstrating GNN learning on real-time data for an anomaly-detection task. Such capabilities reshape the depth and nature of knowledge that can be captured during model training, enabling future integration of GNNs within clinical decision-support systems. Furthermore, the use of synthetic and de-identified data mirrors the privacy concerns inherent to FHIR streams under governance regulations, notably the European Union’s General Data Protection Regulation (GDPR).
References
1. Boll, H. O., Amirahmadi, A., Ghazani, M. M., de Morais, W. O., de Freitas, E. P., Soliman, A., Etminani, F., Byttner, S., & Recamonde-Mendoza, M. (2024). Graph neural networks for clinical risk prediction based on electronic health records: A survey. Journal of Biomedical Informatics, 151, Article 104616.
2. Chan, T. H., Yin, G., Bae, K., & Yu, L. (2024). Multi-task heterogeneous graph learning on electronic health records. Neural Networks, 180, Article 106644.
3. Cho, H. N., Ahn, I., Gwon, H., Kang, H. J., Kim, Y., Seo, H., Choi, H., Kim, M., Han, J., Kee, G., Jun, T. J., & Kim, Y. H. (2022). Heterogeneous graph construction and HinSAGE learning from electronic medical records. Scientific Reports, 12, Article 21152.
4. Mashetty, S. (2024). Leveraging Deep Learning, Neural Networks, and Data Engineering for Intelligent Mortgage Loan Validation: A Data-Driven Approach to Automating Borrower Income, Employment, and Asset Verification. Neural Networks, and Data Engineering for Intelligent Mortgage Loan Validation: A Data-Driven Approach to Automating Borrower Income, Employment, and Asset Verification (December 21, 2024).
5. D’Souza, N. S., Wang, H., Giovannini, A., Foncubierta-Rodriguez, A., Beck, K. L., Boyko, O., & Syeda-Mahmood, T. F. (2024). Fusing modalities by multiplexed graph neural networks for outcome prediction from medical data and beyond. Medical Image Analysis, 93, Article 103064.
6. Krishnan, R., Rajpurkar, P., & Topol, E. J. (2022). Self-supervised learning in medicine and healthcare. Nature Biomedical Engineering, 6, 1346–1352.
7. Lu, C., Reddy, C. K., & Ning, Y. (2023). Self-supervised graph learning with hyperbolic embedding for temporal health event prediction. IEEE Transactions on Cybernetics, 53(4), 2124–2136.
8. KOLLA, S. K. (2024). DEEP LEARNING AND CLINICAL NLP FRAMEWORK FOR AUTOMATED KNOWLEDGE EXTRACTION FROM MULTIMODAL HEALTH . TPM – Testing, Psychometrics, Methodology in Applied Psychology, 31(S2), 134–140. Retrieved from https://tpmap.org/submission/index.php/tpm/article/view/4748
9. Mishra, R., & Shridevi, S. (2024). Knowledge graph driven medicine recommendation system using graph neural networks on longitudinal medical records. Scientific Reports, 14, Article 25449.
10. Molaei, S., Bousejin, N. G., Ghosheh, G. O., Thakur, A., Chauhan, V. K., Zhu, T., et al. (2024). CliqueFluxNet: Unveiling EHR insights with stochastic edge fluxing and maximal clique utilisation using graph neural networks. Journal of Healthcare Informatics Research, 8, 555–575.
11. Mahadevan, S. (2024). Empowering Manufacturing: Generative AI Revolutionizes ERP Application. Int. J. Innov. Sci. Res, 9, 593-595.
12. Nguyen, T. M., Poh, K. L., Chong, S. L., et al. (2024). Med-MGF: Multi-level graph-based framework for handling medical data imbalance and representation. BMC Medical Informatics and Decision Making, 24, Article 242.
13. Pellegrini, C., Navab, N., & Kazi, A. (2023). Unsupervised pre-training of graph transformers on patient population graphs. Medical Image Analysis, 89, Article 102895.
14. Manukonda, K. R. R., Swamy, H., Sankar, R., & CM, C. (2024, August). Application of Explicit Group Sparse Projection in Optimizing 5G Network Resource Allocation and Topology Design. In 2024 4th Asian Conference on Innovation in Technology (ASIANCON) (pp. 1-6). IEEE.
15. Wang, Y., Sun, Z., He, Q., Li, J., Ni, M., & Yang, M. (2023). Self-supervised graph representation learning integrates multiple molecular networks and decodes gene-disease relationships. Patterns, 4(1), Article 100651.
16. Wang, Y., Zhang, R., Yang, Q., Zhou, Q., Zhang, S., Fan, Y., Huang, L., Li, K., & Zhou, F. (2024). FairCare: Adversarial training of a heterogeneous graph neural network with attention mechanism to learn fair representations of electronic health records. Information Processing & Management, 61(3), Article 103682.
17. Movva, V. R. (2024). Cloud-Native Platform Engineering for High-Scale Retail Systems: Designing Resilient Microservices and Intelligent Infrastructure. Available at SSRN 7449480.
18. Chan, T. H., Yin, G., Bae, K., & Yu, L. (2024). Multi-task heterogeneous graph learning on electronic health records. Neural Networks, 180, Article 106644.
19. Pellegrini, C., Navab, N., & Kazi, A. (2023). Unsupervised pre-training of graph transformers on patient population graphs. Medical Image Analysis, 89, Article 102895.
20. Lu, C., Reddy, C. K., & Ning, Y. (2023). Self-supervised graph learning with hyperbolic embedding for temporal health event prediction. IEEE Transactions on Cybernetics, 53(4), 2124–2136.
21. Manukonda, K. R. R. (2024). Chapter-5 Enhancing Test Automation Efficiency Through AI-driven Reporting: An Empirical Study on Defect Detection and Resolution Time. ADVANCES IN, 119, 107.
22. Dwivedi, V. P., Rampášek, L., Galkin, M., Paranjape, B., Martinkus, K., Azevedo, R. E., & Lió, P. (2022). Graph neural networks with learnable structural and positional representations. International Conference on Learning Representations.
23. Hou, Z., Liu, X., Cen, Y., Dong, Y., Yang, H., Wang, C., & Tang, J. (2022). GraphMAE: Self-supervised masked graph autoencoders. Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 594–604.
24. Hou, Z., Liu, X., Dong, Y., Wang, C., & Tang, J. (2023). GraphMAE2: A decoding-enhanced masked self-supervised graph learner. Proceedings of the ACM Web Conference 2023, 737–746.
25. Rohit Gorle. (2024). Deep Learning Approaches for Threat Detection in Multi-Cloud Environments. Journal of Computational Analysis and Applications (JoCAAA), 33(07), 3857–3864. Retrieved from https://eudoxuspress.com/index.php/pub/article/view/5879
26. Thakoor, S., Tallec, C., Azar, M. G., Munos, R., Veličković, P., & Valko, M. (2021). Bootstrapped representation learning on graphs. NeurIPS 2021 Workshop on Deep Learning on Graphs: Methods and Applications.
27. You, Y., Chen, T., Sui, Y., Chen, T., Wang, Z., & Shen, Y. (2021). Graph contrastive learning with augmentations. Advances in Neural Information Processing Systems, 34.
28. Zhang, Z., Liu, Q., Wang, H., Lu, S., & Lee, C. (2021). Motif-based graph self-supervised learning for molecular property prediction. Advances in Neural Information Processing Systems, 34.
29. Xia, J., Zhu, Y., Du, Y., & Li, S. Z. (2022). A survey of graph neural networks in real-world applications. ACM Computing Surveys.
30. Wu, L., Cui, P., Pei, J., Zhao, L., & Guo, X. (2022). Graph neural networks: Foundations, frontiers, and applications. Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining.
31. Jin, W., Liu, X., Zhao, X., Tang, Y., Shah, N., & Jiang, J. (2022). Automated self-supervised learning for graphs. International Conference on Learning Representations.
32. Sreenivasulu Naidu Yalakaturi. (2024). AI-Assisted Reference Data Harmonization Across Global Data Center Systems. Journal of Computational Analysis and Applications (JoCAAA), 33(08), 9576–9587. Retrieved from https://eudoxuspress.com/index.php/pub/article/view/5790
33. He, X., Bresson, X., Laurent, T., Perold, A., LeCun, Y., & Hooi, B. (2023). Harnessing explanations: LLM-to-graph knowledge distillation for graph learning. Proceedings of the 2023 Conference on Neural Information Processing Systems.
34. Hu, Z., Dong, Y., Wang, K., Chang, K. W., & Sun, Y. (2021). GPT-GNN: Generative pre-training of graph neural networks. Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 1857–1867.
35. Zhang, M., Cai, Z., Zhang, J., & Wang, J. (2022). Self-supervised learning for graph neural networks: A survey. IEEE Transactions on Neural Networks and Learning Systems.
36. Zhang, S., Tong, H., Xu, J., & Maciejewski, R. (2021). Graph convolutional networks: A comprehensive review. Computational Social Networks, 8, Article 1.
37. Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., & Yu, P. S. (2021). A comprehensive survey on graph neural networks. IEEE Transactions on Neural Networks and Learning Systems, 32(1), 4–24.
38. Zhou, J., Cui, G., Hu, S., Zhang, Z., Yang, C., Liu, Z., Wang, L., Li, C., & Sun, M. (2021). Graph neural networks: A review of methods and applications. AI Open, 2, 57–81.
39. Veličković, P. (2023). Everything is connected: Graph neural networks. Current Opinion in Structural Biology, 79, Article 102538.
40. Mahadevan, S., Desani, N. R., Yadav, P. S., Joseph, T., & Malcom, A. (2024). FutureTech Innovations. Cari Journals USA LLC.
41. Li, Y., Tarlow, D., Brockschmidt, M., & Zemel, R. (2021). Gated graph sequence neural networks. International Conference on Learning Representations.
42. Ying, C., Cai, T., Luo, S., Zheng, S., Ke, G., He, D., Shen, Y., & Liu, T. Y. (2021). Do transformers really perform badly for graph representation? Advances in Neural Information Processing Systems, 34.
43. Rong, Y., Bian, Y., Xu, T., Xie, W., Wei, Y., Huang, W., & Huang, J. (2021). Self-supervised graph transformer on large-scale molecular data. Advances in Neural Information Processing Systems, 33, 12559–12571.
44. Chen, D., Lin, Y., Li, W., Li, P., Zhou, J., & Sun, X. (2021). Measuring and relieving the over-smoothing problem for graph neural networks from the topological view. Proceedings of the AAAI Conference on Artificial Intelligence, 34(4), 3438–3445.
Additional Files
Published
Data Availability Statement
The study used synthetic and de-identified FHIR data streams. Patient-identifiable data are not publicly available due to privacy and regulatory restrictions (GDPR, HIPAA).
Issue
Section
License
Copyright (c) 2025 Venkata krishna azith ganti, Daniel Wilson (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under a Creative Commons
Attribution 4.0 International License (CC BY 4.0).
You are free to:
- Share: copy and redistribute the material
- Adapt: remix, transform, and build upon the material
for any purpose, even commercially.
Under the following terms:
Attribution — You must give appropriate credit to
the original author(s) and source.
Full license text: https://creativecommons.org/licenses/by/4.0/
