Journal of clinical monitoring and computing
-
J Clin Monit Comput · Dec 2022
Accuracy of a point-of-care blood lactate measurement device in a prehospital setting.
Point-of-care blood lactate is a promising prognostic biomarker of short-term mortality risk. Portable lactate meters need validation in the prehospital setting before widespread implementation and it is unknown whether the mode of sampling (arterial, capillary or venous) matters. This study aims to compare the StatStrip Xpress Lactate Meter's (SSX) accuracy to a validated blood gas analyser, ABL90 FLEX (ABL90), in arterial samples in the prehospital environment and to determine if lactate levels measured in venous and capillary blood samples are sufficiently accurate compared to arterial lactate levels. ⋯ Bland-Altman plots showed that SSX lactate measurements in arterial, venous and capillary blood samples all had systematically negative biases compared to ABL90. We conclude that the SSX is accurate in our prehospital setting. Venous samples should be preferred over capillary samples, when arterial samples cannot be obtained.
-
J Clin Monit Comput · Dec 2022
EditorialThe impact of eye-tracking on patient safety in critical care.
Patient safety has become a high priority in health care. The recognition, prevention and reduction of human errors are crucial for patient care. ⋯ Diverse studies in critical care have proved the usefulness of eye-tracking to analyze real-life scenarios. These insights could contribute to increased patient safety.
-
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.