Increasing Cardiovascular Data Sampling Frequency and Referencing It to Baseline Improve Hemorrhage Detection.

摘要: We hypothesize that knowledge of a stable personalized baseline state and increased data sampling frequency would markedly improve the ability to detect progressive hypovolemia during hemorrhage earlier with lower false positive rate than when using less granular data. Design Prospective temporal challenge. Setting Large animal research laboratory, University Medical Center. Subjects Fifty-one anesthetized Yorkshire pigs. Interventions Pigs were instrumented arterial, pulmonary central venous catheters allowed stabilize for 30 minutes then bled at constant either 5 mL·min-1 (n = 13) or 20 38) until mean arterial pressure decreased 40 mm Hg in pigs, respectively. Measurements main results Data stabilization period served as baseline. Hemodynamic variables collected 250 Hz used create predictive models "bleeding" featurized beat-to-beat waveform compared unfeaturized hemodynamic averaged over 1-minute simple metrics random forest classifiers identify bleeding without The robustness prediction was evaluated leave-one-pig-out cross-validation. Predictive performance by their activity monitoring operating characteristic receiver profiles. Primary threshold poorly identified bleed onset unless very initial reference available. When referenced baseline, detection rates 10-2 time 80% pigs similar metrics, beat-to-beat, about 3-4 minutes. Whereas universally baselined, increasing reduced latency from 10 8 6 minutes, waveform, Some informative features differed between models. Conclusions Knowledge personal allows early new-onset bleeding, whereas if no exists improves rate.