Artificial Intelligence in Pediatric Blood Transfusion during Anesthesia: A Scoping Review

Parisa  Akbarpour; Parisa  Moradimajd; Azam  Saei; Maryam  Aligholizadeh; Siavash  Sangi

doi:10.18502/aacc.v11i2.17958

Parisa Akbarpour Department of Anesthesia, Faculty of Paramedicine, Iran University of Medical Sciences, Tehran, Iran.
Parisa Moradimajd Department of Anesthesia, Faculty of Paramedicine, Iran University of Medical Sciences, Tehran, Iran.
Azam Saei Department of Anesthesia, Faculty of Paramedicine, Iran University of Medical Sciences, Tehran, Iran.
Maryam Aligholizadeh Department of Anesthesia, Faculty of Paramedicine, Iran University of Medical Sciences, Tehran, Iran.
Siavash Sangi Department of Anesthesia, Faculty of Paramedicine, Iran University of Medical Sciences, Tehran, Iran.

DOI: https://doi.org/10.18502/aacc.v11i2.17958

Keywords: Artificial intelligence; Machine learning; Blood transfusion; Anesthesia; Pediatrics

Abstract

Background: Transfusion is a vital process, but incorrect injection can cause harm. In the field of children's blood transfusion under anesthesia, the use of artificial intelligence (AI) and machine learning (ML) is regarded as innovative tools that enhance patient safety levels. This article examines and reviews the scoping literature on the use of artificial intelligence in pediatric blood transfusions during anesthesia, with the aim of identifying solutions, challenges, and future opportunities in this field.

Methods: The study, conducted from early 2024 to May 2024, aimed to evaluate the effectiveness of artificial intelligence (AI) and machine learning (ML) in predicting blood transfusion needs and bleeding risks in pediatric anesthesia. Relevant keywords, including artificial intelligence, machine learning, predictive model, neural network, predictive algorithm, blood transfusion, children, pediatric, neonates, anesthesia, surgery, and operation, were extracted from the Medical Subject Headings (MeSH). A comprehensive search strategy was independently implemented in Web of Science, PubMed, Scopus, and Google Scholar databases

Results: The search strategy initially identified 260 articles. After a systematic screening process, 60 duplicate articles were excluded. Subsequently, careful screening of titles, abstracts, and full texts eliminated an additional 195 articles, resulting in a final selection of 5 relevant English-language articles. Based on these studies, factors such as the type of surgery, the machine learning models used, decreases in hemoglobin and hematocrit levels before surgery, prolonged surgery, as well as the young age and low weight of pediatric patients, were identified as indicators of the increased risk of blood transfusion during surgery and anesthesia.

Conclusion: Based on the findings of the studies, artificial intelligence (AI) and machine learning (ML) have shown significant advancements in pediatric blood transfusion under anesthesia. This technology offers notable benefits, including high accuracy in predicting transfusion requirements and the ability to make timely decisions in critical situations. However, despite these advancements, further research is warranted to comprehensively understand the advantages and limitations of AI in the field of pediatric blood transfusion during anesthesia.