Enhancing glucose level prediction of ICU patients through hierarchical modeling of irregular time-series.

ABSTRACT: Accurately predicting blood glucose (BG) levels of ICU patients is critical, as both hypoglycemia (BG < 70 mg/dL) and hyperglycemia (BG > 180 mg/dL) are associated with increased morbidity and mortality. This study presents a proof-of-concept machine learning framework, the Multi-source Irregular Time-Series Transformer (MITST), designed to predict BG levels in ICU patients. In contrast to existing methods that rely heavily on manual feature engineering or utilize limited Electronic Health Record (EHR) data sources, MITST integrates diverse clinical data-including laboratory results, medications, and vital signs-without predefined aggregation. The model leverages a hierarchical Transformer architecture, designed to capture interactions among features within individual timestamps, temporal

SUBMITTER: Mehdizavareh H 

PROVIDER: S-EPMC12270796 | biostudies-literature | 2025

REPOSITORIES: biostudies-literature