Prediction of All-cause Mortality with Sepsis-associated Encephalopathy in the ICU Based on Interpretable Machine Learning

Sepsis is the main cause of ICU death and death worldwide, defined as organ failure caused by the hosts uncontrolled immune response to an infection. Sepsis-associated encephalopathy (SAE) is a major comorbidity of sepsis and associated with high mortality and poor long-term prognosis. Most of the current clinical cohort analyses are based on sepsis studies, and prediction and risk analyses for ICU death in SAE patients are rarely reported. At the same time, clinicians rarely focus on the preventive measures and the best management of SAE. We should pay more attention to the worsening outcome of SAE to reduce the occurrence of fatal cases and to anticipate and thus intervene in advance. The purpose of this study is to build interpretable machine learning models to predict the all-cause mortality of SAE after ICU admission and implement the individual prediction and analysis.