Incapacitation Prediction for Readiness in Expeditionary Domains: an Integrated Computational Tool (I-PREDICT)

Injury and incapacitation estimates for combat scenarios are currently educated guesses, at best.  These estimates may be based on simplified injury risk thresholds or on hazard parameters, such as pressure, stress, strain, or force applied to an organ or tissue.

Increasingly, such knowledge is incorporated into computational simulations that can be run repeatedly to explore variations in hazards and physiologic responses in order to assign statistical confidence to predictions of injury risk.  However, current modeling and simulation methods for predicting injury can be inaccurate, predicting regional effects rather than considering the whole-body.  These simulations may not be validated appropriately and may not be based on physiologically- or operationally-relevant loading conditions.

To protect warfighters from injuries, injury prevention standards are needed based on a scientific understanding of hazardous conditions typical of military service and the vulnerability of tissues, organs, and bodily functions to those service hazards.  Such standards will inform the design trade spaces of personal protective equipment (PPE), safer vehicles, and safe-to-operate weapons systems, as well as tactics, techniques, and procedures (TTPs) to protect against injury.  Injury prediction models will also allow improved estimates of casualty types, rates, and severity, which in turn will predict individual and unit readiness during operations and will impact medical treatment requirements.  The development of a highly biofidelic finite element model of the whole human body is needed to inform such applications.

Many of the experimental studies used to parameterize human body computational models have employed cadaveric tissue, which may not adequately represent the biomechanical responses of live human tissue due to donor age, cause of death, post-mortem degradation, altered tissue properties resulting from hypothermic test conditions, and isolation of test samples from surrounding structures found in vivo.  Biofidelic constitutive properties that better represent living human tissue are needed to support model parameterization and improve predictions of injury and functional incapacitation. Efforts at the Office of Naval Research (ONR) are underway to develop new methodologies for more accurate measurements of constitutive properties of human and surrogate (animal model) tissues (both ex vivo and in vivo) across a wide range of strain rates.

Accurate use and calibration of component constitutive models remains an obstacle for predictive modeling due to a lack of reliable sample data. Recent and continued advances in characterization standards and experimental methods are needed to account for material anisotropy, rate dependence, multiphase composition, specimen variability, multiphysics, and multiscale behavior.

The overarching goal of the Incapacitation Prediction for Readiness in Expeditionary Domains: an Integrated Computational Tool (I-PREDICT) program is to provide a TRL 6 in silico “skin-in” integrated finite element computational model of the warfighter’s body to be used for injury prevention and treatment, medical response planning, and equipment design (including tradeoff analysis among design parameters, validation, and testing). I-PREDICT will provide an integrated biomechanical response model of the warfighter using biofidelic constitutive tissue properties and associated pre- and post-processing tools.  This information will be used to predict injury and near term functional incapacitation (reduction in the ability to move, shoot, and/or communicate) in response to specific military hazards, in priority order of: 1) blunt impact/accelerative loading and 2) blast pressure effects.

The I-PREDICT model will be based on experimentally-derived material properties of human tissues at strain rates equivalent to those experienced during military hazards and will be validated with data on regional and whole-body mechanics.  I-PREDICT will include variable anthropometry (e.g., differences in size, weight, somatotype, and age), variable posture, variable biofidelity, and gender differences in modeling.  The results will be incorporated into injury and readiness estimates affecting medical response planning and preliminary design and testing of equipment, resulting in cost savings and more thoroughly vetted products, including principled considerations for engineering tradeoffs (e.g., weight of body armor vs. mobility).

This particular award focused on acquiring a finite element computational model of the thorax to predict injury and functional incapacitation from behind-armor blunt trauma as an initial step toward whole-body modeling of responses to a variety of military hazards.  The research project award recipients were selected from among the offerors responding to MTEC’s Request for Project Proposals (18-04-I-PREDICT).

I-PREDICT Thorax Model Prototype

Project Team: Southwest Research Institute
Award Amount: 3.83M
Project Duration: 48 months
Project Objective: This program is in support of the Office of Naval Research (ONR) Incapacitation Prediction for Readiness in Expeditionary Domains: an Integrated Computational Tool (I-PREDICT) Force Health Protection Future Naval Capability (FNC) project. The overarching objective of the I-PREDICT program is to provide a high fidelity, validated, finite element model of the Warfighter’s body for injury prediction and incapacitation assessment. As a first step towards this goal, the objective of Phase I of the program is to update and quantitatively characterize the predictive performance of a government supplied torso finite element model to predict the risk of injury due to behind armor blunt trauma (BABT) and potential incapacitation. We have implemented a tightly integrated computational/experimental approach that builds upon and leverages the capabilities and expertise of a group of well-established researchers; the assembled team has a documented history of successfully working both together and individually on complex injury biomechanics modeling and experimental characterization programs. This study has been designed to address limitations of the existing government supplied total human body FEM with regard to material property viscoelasticity, high-rate loading, and injury validation using experimental testing and advanced finite element modeling. The project will use data from literature, supplemented with experimental testing in our laboratories to add improved material properties and validate the updated FEM for specific injuries whole system levels using experiments conducted using animal tissues, in vivo porcine specimens, and PMHS. In Phase 2 of the work, the team will: 1) Add a high-fidelity head and cervical spine to the ATBM v2.0 torso model; 2) Characterize the dynamic response of select organ components and quantify the extent of tissue damage/injury; and 3) Develop organ level injury prediction methodologies.

Phase 1 Accomplishments:

  • A tightly coupled experimental and computational approach was implemented to increase the fidelity and validity of the supplied torso finite element model and generate the methodologies necessary to extract understandable and actionable results for the application to behind armor blunt trauma (BABT).
  • Significant updates were implemented into the model including corrections to the mesh, integration of a rigid spine and deformable rib cage, and general parameter updates aimed at increasing the stability of the model.
  • Additional models were created by morphing the original model to the anatomy of a 50% male and 5% female, which allowed investigation of injury risk across different populations.
  • Material properties were also updated using a combination of data from literature and data generated specifically for BABT relevant loading conditions. Data was collected for a variety of tissues including: heart, lung, liver, spleen, kidney, adipose, cartilage, stomach, and intercostal muscle utilizing multiple methodologies.  High rate data was collected using a Split-Hopkinson pressure bare to generate stress-strain data at rates relevant to BABT impacts.  Additionally, stress relaxation data was collected to determine the viscoelastic properties and fit them to a quasi-linear viscoelastic material model.  Once updated, the model was compared to full body cadaver tests that simulated BABT impacts to validate the overall response of the model.
  • The updated BABT torso model was implemented within a probabilistic framework which allows investigation into the effect population variability has on predicted injury risk.  A new methodology was developed for injury prediction calculation on an element by element basis.  This new methodology, while computationally intensive, facilitates a deeper dive into the model results and helps translate them into a “so what” answer that can be communicated to non-technical personnel.

Phase 2 Accomplishments:

  • A head and neck model was integrated onto the Advanced Total Body Model (ATBM) torso to improve inertial characteristics of the model, as well as advance the model to be used in different applications incorporating spinal injury (i.e. flight ejections, long-term flight loading, etc.).
  • A majority of the organ components and tissues in the model were remeshed using hexahedral elements to significantly reduce element count. As such, an 88% element reduction was achieved while maintaining high element quality.
    • The liver, spleen, lungs, heart, flesh, kidneys, and ribcage were remeshed.
    • Additionally, several other organs and internal structures were converted to hollow control volumes.
  • Material characterization was performed for porcine liver, heart, spleen, and lungs using stress relaxation tests in shear and compression.
  • Material characterization was performed of PMHS for both frozen-then-thawed and fresh liver, heart, spleen, and lungs using stress relaxation tests in shear and compression.
  • Several constitutive material models were investigated for the liver, and a two-term Ogden rubber constitutive model was selected to best represent the experimental material characterization data. A limitation was identified in LS-DYNA’s material library when characterizing soft tissue.
    • Using the two-term Ogden rubber, the liver demonstrated computational stability during impact simulations.
  • In preparation for other model applications, such as long-term flight loading and injury, impact simulations of the liver were performed in FEBio to demonstrate feasibility of the open-source code compared to the gold-standard of LS-DYNA. Results were in good agreement between the two software.
  • The cervical spine was updated by remeshing facet joints and intervertebral discs with hexahedral elements. Additionally, previously determined material properties for the soft tissues of the cervical spine were transferred to the ATBM.
  • Impact and stress relaxation tests were performed to characterize the mechanical response of ballistic gel used in the experimental impact studies to ensure the computational model was correctly capturing the behavior of the gel encasing the soft tissues (liver).
  • Organ-level impact experiments were performed using pressurized PMHS liver and heart specimens that were designed to develop a better understanding of injury tolerance.
  • Imaging protocols were developed using contrast agents to better characterize injury volume to assist modeling efforts. After whole-organ experimental impacts, the soft tissue was stained with contrast and imaged to reveal the location and extent of injury for comparison against the computational results. These data can assist in developing injury criteria in the ATBM.
  • All proposed hub and ballistic impacts (from literature) have been performed to demonstrate model stability and biofidelity.