Prediction of sepsis within 24 hours at the triage stage in emergency departments using machine learning

BACKGROUND:Sepsis is one of the main causes of mortality in intensive care units(ICUs).Early prediction is critical for reducing injury.As approximately 36%of sepsis occur within 24 h after emergency department(ED)admission in Medical Information Mart for Intensive Care(MIMIC-IV),a prediction system for the ED triage stage would be helpful.Previous methods such as the quick Sequential Organ Failure Assessment(qSOFA)are more suitable for screening than for prediction in the ED,and we aimed to fi nd a light-weight,convenient prediction method through machine learning.METHODS:We accessed the MIMIC-IV for sepsis patient data in the EDs.Our dataset comprised demographic information,vital signs,and synthetic features.Extreme Gradient Boosting(XGBoost)was used to predict the risk of developing sepsis within 24 h after ED admission.Additionally,SHapley Additive exPlanations(SHAP)was employed to provide a comprehensive interpretation of the model's results.Ten percent of the patients were randomly selected as the testing set,while the remaining patients were used for training with 10-fold cross-validation.RESULTS:For 10-fold cross-validation on 14,957 samples,we reached an accuracy of 84.1%±0.3%and an area under the receiver operating characteristic(ROC)curve of 0.92±0.02.The model achieved similar performance on the testing set of 1,662 patients.SHAP values showed that the fi ve most important features were acuity,arrival transportation,age,shock index,and respiratory rate.CONCLUSION:Machine learning models such as XGBoost may be used for sepsis prediction using only a small amount of data conveniently collected in the ED triage stage.This may help reduce workload in the ED and warn medical workers against the risk of sepsis in advance.

High temperature performance of coaxial h-BN/CNT wires above 1,000 ℃:Thermionic electron emission and thermally activated conductivity

The development of wires and cables that can tolerate extremely high temperatures will be very important for probing extreme environments, such as in solar exploration, fire disasters, high-temperature materials processing, aeronautics and astronautics. In this paper, a lightweight high-temperature coaxial h-boron nitride (BN)/carbon nanotube (CNT) wire is synthesized by the chemical vapor deposition (CVD) epitaxial growth of h-BN on CNT yarn. The epitaxially grown h-BN acts as both an insulating material and a jacket that protects against oxidation. It has been shown that the thermionic electron emission (1,200 K) and thermally activated conductivity (1,000 K) are two principal mechanisms for insulation failure of h-BN at high temperatures. The thermionic emission of h-BN can provide the work function of h-BN, which ranges from 4.22 to 4.61 eV in the temperature range of 1,306-1,787 K. The change in the resistivity of h-BN with temperature follows the ohmic conduction model of an insulator, and it can provide the “electron activation energy”(the energy from the Fermi level to the conduction band of h-BN), which ranges from 2.79 to 3.08 eV, corresponding to a band gap for h-BN ranging from 5.6 to 6.2 eV. However, since the leakage current is very small, both phenomena have no obvious influence on the signal transmission at the working temperature. This lightweight coaxial h-BN/CNT wire can tolerate 1,200 ℃ in air and can transmit electrical signals as normal. It is hoped that this lightweight high-temperature wire will open up new possibilities for a wide range of applications in extreme high-temperature conditions.