An official website of the United States government
This paper presents a novel method for predicting hemodynamic responses in hemorrhage resuscitation. The proposed approach, namely, robust nonlinear state space modeling (RNSSM), aims to overcome challenges of identifying reliable models using limited and noisy critical care data by innovatively integrating autoencoder learning and variational Gaussian inference in a unified framework. Simulation results demonstrate the initial feasibility and performance evidence of the RNSSM approach as a digital twin of an animal study in hemorrhage resuscitation scenarios.
more »
« less
Mathematical Modeling of Hemodynamic Responses to Burn Injury and Resuscitation
Fluid resuscitation is an integral part of critical care for burn injury patients where the necessary infusion rate is determined based on patient’s urinary output (UO). Motivated by an increasing interest in model-based development and in silico testing of automated burn resuscitation algorithms, we are investigating mathematical modeling of hemodynamic responses to burn injury and resuscitation. The model consists of 3 main components: (1) multi-compartmental volume kinetics including vascular and interstitial fluids and the associated flow interactions, (2) burn-induced hemodynamic perturbation including alterations in tissue permeability and compliance as well as denaturation with protein release, and (3) renal regulatory function including glomerular filtration rate as a function of intravascular volume state and reabsorption function representing the UO dependence on vasopressin. Preliminary evaluation of the initial model with data collected from animals show that the model can reproduce general trend of hemodynamic responses anticipated from burn injury and resuscitation.
more »
« less
- Award ID(s):
- 1748762
- PAR ID:
- 10081554
- Date Published:
- Journal Name:
- Carnegie Mellon Forum on Biomedical Engineering
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract This paper concerns the design and rigorous in silico evaluation of a closed-loop hemorrhage resuscitation algorithm with blood pressure (BP) as controlled variable. A lumped-parameter control design model relating volume resuscitation input to blood volume (BV) and BP responses was developed and experimentally validated. Then, three alternative adaptive control algorithms were developed using the control design model: (i) model reference adaptive control (MRAC) with BP feedback, (ii) composite adaptive control (CAC) with BP feedback, and (iii) CAC with BV and BP feedback. To the best of our knowledge, this is the first work to demonstrate model-based control design for hemorrhage resuscitation with readily available BP as feedback. The efficacy of these closed-loop control algorithms was comparatively evaluated as well as compared with an empiric expert knowledge-based algorithm based on 100 realistic virtual patients created using a well-established physiological model of cardiovascular (CV) hemodynamics. The in silico evaluation results suggested that the adaptive control algorithms outperformed the knowledge-based algorithm in terms of both accuracy and robustness in BP set point tracking: the average median performance error (MDPE) and median absolute performance error (MDAPE) were significantly smaller by >99% and >91%, and as well, their interindividual variability was significantly smaller by >88% and >94%. Pending in vivo evaluation, model-based control design may advance the medical autonomy in closed-loop hemorrhage resuscitation.more » « less
-
Globally, traumatic injury is a leading cause of suffering and death. The ability to curtail damage and ensure survival after major injury requires a time‐sensitive response balancing organ perfusion, blood loss, and portability, underscoring the need for novel therapies for the prehospital environment. Currently, there are few options available for damage control resuscitation (DCR) of trauma victims. We hypothesize that synthetic polymers, which are tunable, portable, and stable under austere conditions, can be developed as effective injectable therapies for trauma medicine. In this work, we design injectable polymers for use as low volume resuscitants (LVRs). Using RAFT polymerization, we evaluate the effect of polymer size, architecture, and chemical composition upon both blood coagulation and resuscitation in a rat hemorrhagic shock model. Our therapy is evaluated against a clinically used colloid resuscitant, Hextend. We demonstrate that a radiant star poly(glycerol monomethacrylate) polymer did not interfere with coagulation while successfully correcting metabolic deficit and resuscitating animals from hemorrhagic shock to the desired mean arterial pressure range for DCR – correcting a 60% total blood volume (TBV) loss when given at only 10% TBV. This highly portable and non‐coagulopathic resuscitant has profound potential for application in trauma medicine.more » « less
-
Fluid therapy is a common treatment for hypovolemic scenarios to restore the lost blood volume and stabilize acutely ill patients. Automating fluid therapy can lead to a reduction of delay in care, a decrement in dosing errors, and a reduction of cognitive load on clinicians responsible for patient care resulting in improved patient outcomes. However, this process is highly challenging due to the complexity of patient’s physiology and the variability of hemodynamic responses among patients. This work presents a novel machine learning approach based on reinforcement learning (RL) for automated fluid management, where the RL agent is designed to recommend subject-specific infusion dosages without having the knowledge of dose-response models and only by interacting with the environment (virtual subject generator). Compared to the state-of-the-art focusing on the entire population’s data, the proposed approach uses individual patient’s data to recommend patient-specific fluid dosage adjustment. Simulation results demonstrate that the proposed approach outperforms a proportional-integral-derivative (PID) and a rule-based fluid resuscitation controller previously reported for an animal study.more » « less
-
Abstract Wildfires have the potential to dramatically alter the carbon (C) storage potential, ecological function, and the fundamental mechanisms that control the C balance of Pacific Northwest (PNW) forested ecosystems. In this study, we explored how wildfire influences processes that control soil C stabilization and the consequent soil C persistence, and the role of previous fire history in determining soil C fire response dynamics. We collected mineral soils at four depth increments from burned (low, moderate, and high soil burn severity classes) and unburned areas and surveyed coarse woody debris (CWD) in sites within the footprint of the 2020 Holiday Farm Fire and in surrounding Willamette National Forest and the H.J. Andrews Experimental Forest. We found few changes in overall soil C pools as a function of fire severity; we instead found that unburned sites contained high levels of pyrogenic C (PyC) that were commensurate with PyC concentrations in the high severity burn sites—pointing to the high background rate of fire in these ecosystems. An analysis of historical fire events lends additional support, where increasing fire count is loosely correlated with increasing PyC concentration. An unexpected finding was that PyC concentration was lower in low soil burn severity sites than in control sites, which we attribute to fundamental ecological differences in regions that repeatedly burn at high severity compared with those that burn at low severity. Our CWD analysis showed that high mean fire return interval (decades between fire events) was strongly correlated with low annual CWD accumulation rate; whereas areas that burn frequently had a high annual CWD accumulation rate. Within the first year postfire, trends in soil density fractions demonstrated no significant response to fire for the mineral-associated organic matter pool but slight increases in the particulate pool with increasing soil burn severity—likely a function of increased charcoal additions. Overall, our results suggest that these PNW forest soils display complex responses to wildfire with feedbacks between CWD pools that provide varying fuel loads and a mosaic fire regime across the landscape. Microclimate and historic fire events are likely important determinants of soil C persistence in these systems.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
