Digital Technologies and Artificial Intelligence in the Diagnosis, Monitoring, and Prevention of Thalassemia: Current Status, Limitations, and Future Perspectives

Abstract

Thalassemia remains a major global health burden requiring timely diagnosis, long-term follow-up, and comprehensive management. In recent years, digital technologies and artificial intelligence (AI) have increasingly been considered as tools that may complement existing approaches to screening, differential diagnosis, complication monitoring, and clinical decision support. This narrative review summarizes current evidence on the application of AI and digital technologies in thalassemia. It examines studies addressing the use of machine learning and deep learning algorithms for the differential diagnosis of thalassemia and iron deficiency anemia, carrier detection, medical image analysis, prediction of selected clinical outcomes, and telemedicine-based monitoring. Particular attention is given to the potential role of AI in personalizing transfusion and chelation therapy, as well as in supporting decision-making in complex clinical situations, including the assessment of candidates for hematopoietic stem cell transplantation. Possible applications of AI in prevention programs, genetic screening, and counseling are also discussed. Alongside potential benefits, this review analyzes key limitations to AI implementation, including data quality and representativeness, model generalizability, interpretability, ethical and regulatory concerns, and infrastructural barriers. Current evidence suggests that AI has promising applications in thalassemia; however, the clinical maturity of different approaches varies considerably, and broader implementation will require further validation, standardization, and evaluation in real-world settings. AI has considerable potential to strengthen thalassemia care, but meaningful implementation will depend on rigorous validation, interpretability, and integration into real-world clinical systems.