Identification of sepsis-associated encephalopathy biomarkers through machine learning and bioinformatics approaches

Abstract

Sepsis-associated encephalopathy (SAE) is common in septic patients, characterized by acute and long-term cognitive impairment, and is associated with higher mortality. This study aimed to identify SAE-related biomarkers and evaluate their diagnostic potential. We analyzed three SAE-related sequencing datasets, using two as training sets and one as a validation set. Weighted Gene Co-expression Network Analysis and four machine learning methods—Elastic Net regression, LASSO, random forest, and XGBoost—were employed, dentifying 18 biomarkers with significant expression changes. External validation and in vitro experiments confirmed the differential expression of these biomarkers. These findings provide insights into SAE pathogenesis and suggest potential therapeutic targets.