Notice of Intent to Sole Source

This Research and Development support for BHSAI will use the Dephy Exoboot (Dephy Inc., Maynard, MA) lower-body exoskeleton suit used by the U.S. will be assessed. Anthropometric measurements (i.e., the height and body mass) of each participant in the laboratory study will be collected as well as computed tomography (CT) images of the left lower-extremity bones (i.e., the tibia and metatarsal) and the lumbar spine. Next, during the marching experiments, motion-capture data using a system that tracks retroreflective markers on the subject’s body will be collected. In addition, during the experiment, electromyography (EMG) data from major muscle groups in the left lower extremities of each subject will also be collected.

Based on the collected experimental data, individualized musculoskeletal models will be developed by incorporating subject-specific bone geometry, body size, and body motion. The extracted subject-specific bone geometry of the tibia, metatarsal, and lumbar spine from the CT images collected will be used to create the individual-specific bone geometries that replaces the corresponding bones in the generic musculoskeletal model available in the AnyBody modeling system (AnyBody Technology, Aalborg, Denmark). The individualized model to match the anthropometric measurements will be created by scaling the generic to each subject by incorporating subject-specific body height, mass, and fat percentage. The collected motion-capture data will be incorporated into the subject-specific musculoskeletal model to predict the kinematics (i.e., joint angles), kinetics (i.e., joint forces and moments), and muscular activities in the lower extremities and lower back of each subject under different conditions. The validation of the individualized musculoskeletal models will be done by comparing the time courses of the predicted muscle forces to the collected EMG data.

In order to quantify the biomechanical responses and injury risks, subject-specific, three-dimensional FE models of the tibia, metatarsal, and lumbar spine for each subject by using the collected CT images will be developed. This entails using the 3-Matic software (Materialise, Leuven, Belgium) to create tetrahedral meshes for each bone based on the bone geometry extracted from the CT images. Optimization schemes available in 3-Matic needs to be used to ensure the generation of high-quality meshes required for FE simulations. If the element-quality check performed within the Abaqus/Explicit software (Dassault Systèmes, Velizy-Villacoublay, France) fails, HyperMesh software (Altair Inc., Troy, MI) needs to be deployed to identify the faulty mesh elements and change their size and shape by rotating, extending, and morphing them. Next, subject-specific material properties (i.e., the elastic modulus) will be calculated from the CT images and mapped to the subject-specific bone meshes. By coupling the bone meshes with the developed musculoskeletal models, the muscle/ligament insertion points for providing the boundary conditions for the FE model can be identified. Furthermore, the body forces and moments obtained from the musculoskeletal model will be used as subject-specific loading conditions for the FE model. All FE simulations will be performed using Abaqus/Explicit software, using high-end computing resources at the DoD Supercomputing Resource Center at the U.S. Army Research Laboratory.

Based on the calculated individualized biomechanical responses, the risk of stress fracture in the tibia and in the lumbar spine can be estimated by using a previously developed statistical risk-injury model. Briefly, the model used by HJF incorporates the contribution of bone-damage accumulation from repetitive loading, bone-damage removal by adaptive remodeling, as well as bone repair. The model also considers bone-tissue failure due to fatigue by accounting for bone size and number of loading cycles. For estimating stress-fracture risk of the metatarsal and lumbar spine, the existing statistical model will be modified by incorporating the characteristics of these bones (i.e., bone size, loading, and fatigue and repair parameters) into the model. Using the modified model, the stress-fracture risk of the metatarsal and lumbar spine for each subject can then be estimated under different marching conditions. This will then allow statistical analyses that can be used to develop a regression model to stratify the injury risk as a function of load carriage, marching distance, and terrain.