A Comparison of Five Algorithmic Methods and Machine Learning Pattern Recognition for Artifact Detection in Electronic Records of Five Different Vital Signs: A Retrospective Analysis.

BACKGROUND: Research on electronic health record physiologic data is common, invariably including artifacts. Traditionally, these artifacts have been handled using simple filter techniques. The authors hypothesized that different artifact detection algorithms, including machine learning, may be necessary to provide optimal performance for various vital signs and clinical contexts. METHODS: In a retrospective single-center study, intraoperative operating room and intensive care unit (ICU) electronic health record datasets including heart rate, oxygen saturation, blood pressure, temperature, and capnometry were included. All records were screened for artifacts by at least two human experts. Classical artifact detection methods (cutoff, multiples of SD [z-value], interquartile range, and local outlier factor) and a supervised learning model implementing long short-term memory neural networks were tested for each vital sign against the human expert reference dataset. For each artifact detection algorithm, sensitivity and specificity were calculated. RESULTS: A total of 106 (53 operating room and 53 ICU) patients were randomly selected, resulting in 392,808 data points. Human experts annotated 5,167 (1.3%) data points as artifacts. The artifact detection algorithms demonstrated large variations in performance. The specificity was above 90% for all detection methods and all vital signs. The neural network showed significantly higher sensitivities than the classic methods for heart rate (ICU, 33.6%; 95% CI, 33.1 to 44.6), systolic invasive blood pressure (in both the operating room [62.2%; 95% CI, 57.5 to 71.9] and the ICU [60.7%; 95% CI, 57.3 to 71.8]), and temperature in the operating room (76.1%; 95% CI, 63.6 to 89.7). The CI for specificity overlapped for all methods. Generally, sensitivity was low, with only the z-value for oxygen saturation in the operating room reaching 88.9%. All other sensitivities were less than 80%. CONCLUSIONS: No single artifact detection method consistently performed well across different vital signs and clinical settings. Neural networks may be a promising artifact detection method for specific vital signs.

Impact of noninvasive ventilation at a municipal emergency department on ICU admissions.

BACKGROUND: In 2015, the emergency department of a municipal hospital in Vienna began to perform noninvasive ventilation (NIV) on patients admitted for acute respiratory failure, given no intubation criteria were met. The intention of this study was to show to which type of hospital unit patients were transferred after undergoing NIV in the emergency department. Additionally, the impact of the underlying disease, a patient's sex and age and the year of intervention were analyzed. METHODS: A single-center retrospective exploratory study was performed on 371 patients. All patients with acute respiratory failure who were noninvasively ventilated at the study center emergency department from 2015 to 2018 were eligible. Relevant data were extracted from the patient's medical records. RESULTS: A total of 43.7% (95% confidence interval, CI 38.8-48.5%) of patients were successfully stabilized in the emergency department through NIV and subsequently transferred to a normal care unit or discharged. This nonintensive care admission rate was significantly associated with certain underlying medical conditions, age and year of intervention. A further 19.7% (95% CI 15.6-23.7%) of patients were transferred to an intermediate care unit instead of an intensive care unit. CONCLUSION: These findings emphasize the importance of noninvasive ventilation at the emergency department in reducing load on intensive care units and ensuring an efficient hospital workflow. Nonintensive care admission rate appears to be the highest in patients with pulmonary edema, especially in the higher age range and is also associated with the level of staff training. Prospective trials are needed to accurately confirm these correlations.