<p>Artificial Intelligence (AI) has evolved the healthcare system, including transfusion services. This scoping review analyses published original articles on the use of AI in transfusion medicine. A total of 89 articles met the eligibility criteria for the review which were categorized based on the domain of transfusion medicine. Most of the published articles were on the use of AI on transfusion prediction, followed by inventory management. In both these domains, the use of AI ensured better transfusion prediction, timely issue of blood and effective utilisation. Other domains included the use of AI to predict transfusion reactions, component quality and blood typing. Although there was a significant contribution of literature from abroad, studies from India were very sparse. To develop indigenous data, large multi-centric studies are required, which will then help in the implementation of AI in blood centres across India. The decentralised nature of transfusion services and lack of uniform data systems are some of the barriers limiting the conduct of AI based studies in India. Federated data-sharing models offers a promising approach in generating collaborative data while maintaining data privacy and blood centre autonomy.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

AI in Transfusion Medicine: A Scoping Review with Bibliography

  • Priyadarsini Jayachandran Arcot,
  • Najla Haneefa Basheela,
  • Suvro Sankha Datta,
  • Karan Kumar,
  • Suchet Sachdev

摘要

Artificial Intelligence (AI) has evolved the healthcare system, including transfusion services. This scoping review analyses published original articles on the use of AI in transfusion medicine. A total of 89 articles met the eligibility criteria for the review which were categorized based on the domain of transfusion medicine. Most of the published articles were on the use of AI on transfusion prediction, followed by inventory management. In both these domains, the use of AI ensured better transfusion prediction, timely issue of blood and effective utilisation. Other domains included the use of AI to predict transfusion reactions, component quality and blood typing. Although there was a significant contribution of literature from abroad, studies from India were very sparse. To develop indigenous data, large multi-centric studies are required, which will then help in the implementation of AI in blood centres across India. The decentralised nature of transfusion services and lack of uniform data systems are some of the barriers limiting the conduct of AI based studies in India. Federated data-sharing models offers a promising approach in generating collaborative data while maintaining data privacy and blood centre autonomy.