Journal of clinical monitoring and computing
-
J Clin Monit Comput · Oct 2022
Multicenter StudyPredicting hypoglycemia in critically Ill patients using machine learning and electronic health records.
Hypoglycemia is a common occurrence in critically ill patients and is associated with significant mortality and morbidity. We developed a machine learning model to predict hypoglycemia by using a multicenter intensive care unit (ICU) electronic health record dataset. Machine learning algorithms were trained and tested on patient data from the publicly available eICU Collaborative Research Database. ⋯ The best-performing predictive model was the eXtreme gradient boosting model (XGBoost), which achieved an area under the received operating curve (AUROC) of 0.85, a sensitivity of 0.76, and a specificity of 0.76. The machine learning model developed has strong discrimination and calibration for the prediction of hypoglycemia in ICU patients. Prospective trials of these models are required to evaluate their clinical utility in averting hypoglycemia within critically ill patient populations.
-
J Clin Monit Comput · Oct 2022
Multicenter StudyComparison of a new EMG module, AF-201P, with acceleromyography using the post-tetanic counts during rocuronium-induced deep neuromuscular block: a prospective, multicenter study.
Recent advances in neuromuscular monitors have facilitated the development of a new electromyographic module, AF-201P™. The purpose of this study was to investigate the relationship between post-tetanic counts (PTCs) assessed using the AF-201P™ and the acceleromyographic TOF Watch SX™ during rocuronium-induced deep neuromuscular block. Forty adult patients consented to participate in this study. ⋯ Regression analysis showed no significant difference in PTCs between the two monitors (PTCs measured by the TOF Watch SX™ = 0.78·PTCs measured by AF-201P™ + 0.21, R = 0.56). Bland-Altman analysis also showed acceptable ranges of bias [95% CI] and limits of agreement (0.3 [0.2 to 0.5] and - 4.6 to 5.3) for the PTCs. The new EMG module, AF-201P™, showed reliable PTCs during deep neuromuscular block, as well as the TOF Watch SX™.