Abstract

Rare diseases affect millions of people worldwide. They are characterised by prolonged diagnostic uncertainty, fragmented care pathways and significant logistical and informational barriers. Current hospital information systems and conventional artificial intelligence (AI) approaches are primarily designed for common conditions and remain poorly adapted to the sparse, heterogeneous and longitudinal nature of rare disease data. This article critically reviews recent advances in AI applied to rare diseases, including electronic health record (EHR) analysis, anomaly detection, multimodal phenotyping, large language models (LLMs) and agent-based systems. Based on these developments, the paper presents a practitioner–patient framework centred on the continuous identification of unusual data, rather than direct automated diagnosis. This framework combines anomaly detection models, natural language processing, phenotypic matching based on the Human Phenotype Ontology (HPO), multimodal data integration and explainable AI mechanisms, supporting clinical vigilance throughout the care pathway. In parallel, a secure, multimodal AI assistant is introduced to empower patients by simplifying medical information, improving coordination and reducing administrative burdens. Unlike existing approaches, which focus primarily on early diagnosis, the proposed framework aims to minimise bottlenecks throughout the patient journey, while ensuring human clinical oversight is maintained and compliance with emerging regulatory requirements, such as the GDPR and the EU AI Act, is achieved. The article discusses the technical, ethical, organisational and regulatory challenges associated with such systems, and highlights the potential of AI to improve the timeliness of diagnoses, care coordination and patient autonomy in the context of rare diseases.

Keywords: rare diseases; artificial intelligence; electronic health records; anomaly detection; multimodal phenotyping; large language models; explainable AI; patient pathway optimisation; clinical decision support; healthcare interoperability

References

1. AlSaad R., et al. “Temporal modelling of irregular clinical trajectories in electronic health records for rare diseases”. Artificial Intelligence in Medicine (2024).
2. Boardman-Pretty F., et al. “Evaluating algorithmic approaches to rare disease case-finding: a retrospective validation study using electronic health records”. Orphanet Journal of Rare Diseases (2026).
3. Digitale J., et al. “Methods for addressing missingness in electronic health records when building clinical prediction models”. JMIR Medical Informatics (2025).
4. Doan LMT., et al. “Bridging modalities with AI: a review of AI advances in multimodal biomedical imaging”. Commun Eng 5 (2026): 30.
5. Faviez C., et al. “Enhancing rare disease detection with deep phenotyping from EHR narratives: evaluation on Jeune syndrome”. International Journal of Medical Informatics 203 (2025): 106021.
6. Gebeyehu B., et al. “Detection of rare medical events in electronic health records using machine learning: Current practices and suggestions – A scoping review”. PLOS ONE (2026).
7. Germain DP. “Applying artificial intelligence to rare diseases: a literature review highlighting lessons from Fabry disease”. Orphanet J Rare Dis 20 (2025): 186.
8. Greco KF., et al. “A weakly supervised transformer for rare disease diagnosis and subphenotyping from EHRs with pulmonary case studies”. npj Digital Medicine (2026).
9. Groza T, Baynam G and Jamuar SS. “Reimagining care of people living with rare diseases with artificial intelligence”. PLOS Medicine 23.2 (2026): e1004966.
10. Kolekar SS., et al. “AI Systems for Diagnosing Rare Diseases: Challenges and Solutions”. Conference: ICIMMI 2024: 6th international conference on information management & machine intelligence (2024).
11. Lin S., et al. “An artificial intelligence-based approach for identifying patients with rare diseases from electronic health records: An artificial intelligence-based approach for identifying patients with rare diseases using retrospective electronic health records applied to Pompe disease”. Frontiers in Neurology (2023).
12. Mao X., et al. “A phenotype-based AI pipeline outperforms human experts in differentially diagnosing rare diseases using EHRs”. npj Digital Medicine (2025).
13. Patharkar A., et al. “Predictive modelling of biomedical temporal data in healthcare applications: review and future directions”. Front. Physiol 15 (2024).
14. Salmi M., et al. “Handling imbalanced medical datasets: review of a decade of research”. Artificial Intelligence Review 57 (2024): 273.
15. Seifi S., et al. “Complying with the EU AI Act: Innovations in explainable and user-centric hand gesture recognition”. Machine Learning with Applications 20 (2025): 100655.
16. Shyr C., et al. “Large Language Models for Rare Disease Diagnosis at the Undiagnosed Diseases Network”. JAMA Network Open 8.8 (2025): e2528538.
17. Sreenivasan M., et al. “Temporal and comorbidity-aware representation of longitudinal patient trajectories from electronic health records”. Biomedical Physics & Engineering Express 12.1 (2026).
18. Visibelli A., et al. “The Impact of Artificial Intelligence in the Odyssey of Rare Diseases”. Journal of Personalized Medicine (2023).
19. Viswan V., et al. “Multimodal fusion and explainability of artificial intelligence models in Alzheimer’s Disease detection”. Brain Informatics 13.1 (2026): 5.
20. Xin Wang, Da He and Chunlin Jin. “AI-driven enhancements in rare disease diagnosis and support system optimization”. Intractable Rare Dis Res 14.4 (2025): 306-308.
21. Magda Wojtara., et al. “Artificial intelligence in rare disease diagnosis and treatment”. Clin Transl Sci 16.11 (2023): 2106-2111.
22. Zhao W., et al. “An agentic system for rare disease diagnosis with traceable reasoning”. Nature (2026).
23. Zhu M., et al. “Bridging data gaps of rare conditions in ICU: a multi-task domain adaptation approach”. npj Digital Medicine (2026).

Citation

Robin Vivian. “Artificial Intelligence to Help Reduce Delays and Bottlenecks in the Care Pathways for Rare Diseases: An Integrated Practitioner-Patient Approach". Clareus Scientific Science and Engineering 3.3 (2026): 15-27.

Copyright

© 2026 Robin Vivian. Licensee Clareus Scientific Publications. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license.