Understanding the risk factors for adverse events during exchange transfusion in neonatal hyperbilirubinemia using explainable artificial intelligence

Abstract

Objective To understand the risk factors associated with adverse events during exchange transfusion (ET) in severe neonatal hyperbilirubinemia. Study design We conducted a retrospective study of infants with hyperbilirubinemia who underwent ET within 30 days of birth from 2015 to 2020 in a children’s hospital. Both traditional statistical analysis and state-of-the-art explainable artificial intelligence (XAI) were used to identify the risk factors. Results A total of 188 ET cases were included; 7 major adverse events, including hyperglycemia (86.2%), top-up transfusion after ET (50.5%), hypocalcemia (42.6%), hyponatremia (42.6%), thrombocytopenia (38.3%), metabolic acidosis (25.5%), and hypokalemia (25.5%), and their risk factors were identified. Some novel and interesting findings were identified by XAI. Conclusions XAI not only achieved better performance in predicting adverse events during ET but also helped clinicians to more deeply understand nonlinear relationships and generate actionable knowledge for practice.

Big Data in Transfusion Medicine and Artificial Intelligence Analysis for Red Blood Cell Quality Control

“…Although intensively discussed, the microfluidic-driven, image-based judgment of RBC quality is not the only possible application of AI in transfusion medicine (e.g., [24, 99, 100]). However, so far, all these approaches have in common that they are only used at the academic level or in clinical trials without routine application in clinical laboratories.…”

Machine learning in transfusion medicine: A scoping review

“…While the focus of ML is on prediction, and a causal relationship cannot be assumed of the covariates found to have high predictive value, identification of novel risk factors for hypothesis generation and further research can be useful as seen in transfusion-associated lung injury (TRALI) 85 and in pediatric transfusion-associated hyperkalemia. 86 Recognizing that transparency and accountability are essential for clinicians in generating hypotheses, Zhu et al 87,88 focus on explainable AI when presenting adverse events during neonatal hyperbilirubinemia exchange transfusion, particularly through use of SHapley Additive exPlanation (SHAP).…”