Fig. 2
Hemodynamic data captured during lower body negative pressure (LBNP) and release. Measures from A–D are synchronously captured. Each faint line represents a single protocol, while the emboldened line represents the average of all protocols. A Stroke volume (SV) percent change during progressively severe LBNP (i.e., hemorrhage model) and release of LBNP (i.e., rapid transfusion model). B Mean arterial pressure (MAP) percent change. C velocity time integral (VTI) from the wearable Doppler percent change. D corrected flow time (ccFT) percent change. E The optimal carotid artery maximum VTI threshold for distinguishing ≥ + 10% SV∆. Each data point represents a 10-s average. Prior to subsampling, there were 3596 data points categorized as < + 10% SV∆ and 598 data points categorized as ≥ + 10% SV∆. The data categorized as < + 10% SV∆ were randomly subsampled, iteratively 1000 times, to 598 data points (see methods). The sensitivity of maximum VTI is 532/598 = 89% and specificity is 598/598 = 100%. F The optimal ccFT threshold for distinguishing + 10% SV∆. Each data point represents a 10-s average. The sensitivity is 532/598 = 89% and specificity is 598/598 = 100%