The Strategy of Blood Supply and Demand Management and The Application of Artificial Intelligence Models in Emergencies

Vollständiger Abstract

Worum geht es in dieser Arbeit?

As critical and unique emergency supplies, blood resources are characterized by time sensitivity and demanding preservation requirements. In recent years, emergencies such as natural disasters, conflicts, and epidemics have become increasingly frequent worldwide. These events have not only caused widespread social and economic disruption but have also destabilized the blood supply and caused persistent and severe blood shortages. When an emergency occurs, blood services must assess needs and plan actions proactively. Artificial intelligence models can capture higher-order relationships between covariates and outcomes in large datasets and play an important role in addressing blood supply and demand challenges. The use of appropriate artificial intelligence (AI) models to predict and configure the demand and supply of clinical blood products is therefore essential for assisting blood service planning in emergencies. However, a comprehensive review of the application value of AI models in emergency strategies for blood supply and demand management remains lacking. This work summarizes commonly used prediction models for blood demand, strategies for blood supply and demand management in emergencies, and model applications, with the goal of helping hospitals and blood centers select effective response strategies through different predictive models to maximize the protection of patients' lives and health. This review clarifies emergency blood management strategies for different types of emergencies, outlines core selection criteria for AI models, recommends time series models for linear and stationary data, and machine learning models for complex non-linear data, and offers practical guidance for emergency blood supply assurance.