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. Forty-four features including patient demographics, laboratory test results, medications, and vitals sign recordings were considered. The outcome of interest was the occurrence of a hypoglycemic event (blood glucose?<?72 mg/dL) during a patient’s ICU stay. Machine learning models used data prior to the second hour of the ICU stay to predict hypoglycemic outcome. Data from 61,575 patients who underwent 82,479 admissions at 199 hospitals were considered in the study. 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.
相似文献Context:
Wound measurement is an important aspect of wound management. Though there are many techniques to measure wounds, most of them are either cumbersome or too expensive.Aims:
To introduce a simple and accurate technique by which wounds can be accurately measured.Settings and Design:
This is a comparative study of 10 patients whose wounds were measured by three techniques, i.e. ruler, graph and our technique.Materials and Methods:
The graph method was taken as the control measurement. The extent of deviation in wound measurements with our method was compared with the standard technique. The statistical analysis used was ANOVA.Results:
The ruler method was highly inaccurate and overestimated the wound size by nearly 50%. Our technique remained consistent and accurate with the percentage of over or underestimation being 2-4% in comparison with the graph method.Conclusions:
This technique is simple and accurate and is an inexpensive and non-invasive method to accurately measure wounds.KEY WORDS: Digital planimetry, photography, wound measurementYou can’t manage what you can’t measureW. Edwards Deming相似文献