Publications

Massive hemorrhage remains the primary cause of potentially preventable death in traumatic injuries. Monitoring hemorrhage and resuscitation accurately can improve outcomes but continues to be challenging since traditional vital signs are highly compensated by the body. Previous work has developed physiologically interpretable algorithms to assess volume status in simulated models of hemorrhage and resuscitation. In this paper, we further develop these algorithms to assess volume status in a nonhuman primate model of controlled blood loss and subsequent whole-blood resuscitation. We acquired arterial blood pressure (ABP) waveform data on 12 adult male baboons during a step-and-hold protocol for hemorrhage and a constant resuscitation rate. A gradient-boosted regression tree model trained on only the ejected-wave pulse area (EWPA) feature yielded a 19% root-mean-square-error (RMSE), 0.71 R2, and an area under the receiver operating characteristic curve of ≥ 0.9 for key operating points of volume status. The performance of this model with a single feature compares well to results reported previously from single-feature machine-learning (ML) models as well as more complex machine learning models that are difficult to interpret and computationally intensive. This study is the first investigation of these physiologically interpretable models on invasively measured ABP waveforms.

Hemorrhage is the leading cause of preventable death from trauma. Accurate monitoring of hemorrhage and resuscitation can significantly reduce mortality and morbidity but remains a challenge due to the low sensitivity of traditional vital signs in detecting blood loss and possible hemorrhagic shock. Vital signs are not reliable early indicators because of physiological mechanisms that compensate for blood loss and thus do not provide an accurate assessment of volume status. As an alternative, machine learning (ML) algorithms that operate on an arterial blood pressure (ABP) waveform have been shown to provide an effective early indicator. However, these ML approaches lack physiological interpretability. In this paper, we evaluate and compare the performance of ML models trained on nine ABP-derived features that provide physiological insight, using a database of 13 human subjects from a lower-body negative pressure (LBNP) model of progressive central hypovolemia and subsequent progressive restoration to normovolemia (i.e., simulated hemorrhage and whole blood resuscitation). Data were acquired at multiple repressurization rates for each subject to simulate varying resuscitation rates, resulting in 52 total LBNP collections. This work is the first to use a single ABP-based algorithm to monitor both simulated hemorrhage and resuscitation. A gradient-boosted regression tree model trained on only the half-rise to dicrotic notch (HRDN) feature achieved a root-mean-square error (RMSE) of 13%, an R2 of 0.82, and area under the receiver operating characteristic curve of 0.97 for detecting decompensation. This single-feature model's performance compares favorably to previously reported results from more-complex black box machine learning models. This model further provides physiological insight because HRDN represents an approximate measure of the delay between the ABP ejected and reflected wave and therefore is an indication of cardiac and peripheral vascular mechanisms that contribute to the compensatory response to blood loss and replacement.