Artificial Intelligence Applications in Haemophilia Care: A Narrative Review of the Literature

Abstract

INTRODUCTION: Haemophilia is a rare X-linked bleeding disorder caused by deficiencies in coagulation factors, leading to recurrent bleeding episodes, particularly in joints and muscles. Haemophilia A accounts for 80%-85% of cases, while Haemophilia B represents 15%-20%. Despite advances in treatment, challenges such as inhibitor development, treatment variability, data scarcity, algorithmic bias, and disparities in technology access persist. Artificial intelligence (AI) has the potential to improve diagnostic accuracy, prognostication, and management, advancing personalised treatment strategies. AIM: This review examines AI applications in haemophilia care, assessing their impact on diagnosis, predictive modelling, digital health solutions, and treatment optimisation while addressing limitations and ethical concerns. METHODS: A narrative review of 40 articles was conducted, focusing on AI-driven diagnostic tools, predictive modelling, digital health technologies, and treatment optimisation. Additionally, barriers to AI integration, including algorithmic bias, cost, and accessibility, were evaluated. RESULTS: AI enhances diagnostic accuracy, predicts disease severity, assesses inhibitor risks, and optimises recombinant therapies. Machine learning improves precision in robot-assisted surgeries, while AI-powered digital tools, including chatbots and wearables, support self-management and real-time monitoring. Generative AI facilitates patient education and predictive modelling, aiding clinical decision-making. AI-driven individualised prophylaxis strategies using factor mimetics and rebalancing agents are emerging. CONCLUSION: AI represents a paradigm shift toward precision medicine in haemophilia care. However, ethical concerns, data scarcity, and financial barriers limit its full potential. Future research should focus on mitigating biases, improving data availability, and refining AI-driven personalised treatment strategies to optimise patient outcomes.