Artificial intelligence is increasingly being used in transfusion medicine, a field that helps diagnose and manage blood-related conditions including fetal and neonatal alloimmune thrombocytopenia (FNAIT), according to a study published recently in Vox Sanguinis.
Researchers reported that clinicians and scientists showed strong interest in these tools, with many already using them to support clinical care, research and education. Still, experts say broader adoption will require better access to technology, trained professionals and clear rules for safe use.
The findings come from a mixed-methods study that surveyed members of the International Society of Blood Transfusion between June and November 2024. Of 245 initial responses, researchers removed 22 duplicates, leaving 218 participants from 67 countries for analysis.
About 42.7% of respondents were based in Europe. Most respondents were women aged 35 to 55 years and more than half were hematologists, hematopathologists, transfusion medicine physicians or blood bank medical directors.
Overall, 43.5% of respondents said they personally use artificial intelligence tools, and 88.6% had begun using them within the past two years. Those working in academic or university-affiliated centers reported higher adoption than those in nonacademic settings, with 52.4% using AI compared with 32.1% in other workplaces.
Read more about the care team in FNAIT
Professionals in privately funded institutions were also more likely to report using AI than those in government-funded facilities, at 60.5% versus 40.4%. Researchers did not observe significant differences in AI use based on age or gender.
Among clinicians who use AI, 91.1% reported using generative AI programs, and 82.3% said they learned to use them on their own. More than half, 54.4%, applied these tools directly to clinical transfusion medicine tasks such as antibody identification, crossmatch interpretation, classification of transfusion reactions and predicting severe bleeding. These activities can be important when evaluating risks and managing patients affected by conditions such as FNAIT, which occurs when a mother’s immune system attacks her baby’s platelets.
Most users said AI had a positive effect on their work. Overall, 87.3% reported improvements in day-to-day tasks. At the institutional level, 32.7% said their workplaces had integrated AI, and 71.1% of those institutions used it in education, research or both. Educational uses included preparing lectures and generating questions for students. In research, AI helped with brainstorming ideas, statistical analysis, coding, interpreting data and drafting manuscripts.
“Further research on the applications of AI in TM [transfusion medicine] and vendor–academia collaboration may be increasingly required to enable future adoption and workforce development,” explained this study’s authors.
Despite the promise, several barriers remain. A lack of access to AI tools or expertise was cited by 78% of respondents, followed by cost at 74%, difficulty hiring trained professionals at 73% and concerns about data privacy at 72%. Interview participants also raised worries about accuracy, regulatory limits, bias and the possibility that overreliance on AI could affect human judgment or jobs.
The authors said expanding education, building trustworthy clinical tools and strengthening oversight will be key steps to safely integrating artificial intelligence into transfusion medicine and improving care for patients with complex conditions such as FNAIT.
Sign up here to get the latest news, perspectives, and information about FNAIT sent directly to your inbox. Registration is free and only takes a minute.
