Investigating Fall Mitigation Strategies when Walking with an Exoskeleton for Users with Lower-Limb Paralysis

Paralysis as a result of a spinal cord injury can severely impact activities of daily living, limit independence, and negatively impact overall health of Veterans. Powered exoskeletons are an evolving technology aiming to improve mobility and quality of life for individuals with paralysis. Clinical studies of users

walking with exoskeletons report a reduction in pain and spasticity and improvements to bone density and muscle tone. Currently, commercial exoskeletons have FDA 510(k) clearances that require continuous contact guarding or stand-by-assistance by a spotter. During instances of falling, these devices are only designed to enter

joint-locking configurations and require the user to balance themselves via crutches, a walker, or intervention from a caregiver to prevent impact with the ground. The rationale of this CDA-1 is that the absence of safety features that prevent falling while wearing an exoskeleton may be restricting the technology to monitored environments with

trained caregivers and be preventing their widespread use within the community. The purpose of this award is to provide Dr. Hnat with the required training to become a successful researcher within the VA system and to lead her towards achieving her long term goal of designing an exoskeleton that detects destabilizations and

prevents falls from occuring. The current proposal addresses the candidate’s short term goals regarding the issue of falling with powered exoskeletons by exploring configurations to reduce the impact with the ground and the potential of injury should a fall occur. The proposed work is immediately translatable into a hybrid neuroprosthesis

resulting from the primary mentor’s funded research (Dr. Ronald Triolo) and will serve as preliminary data for subsequent CDA-2 and VA Merit Award proposals. The project goals will be achieved through two Specific Aims. Specific Aim 1 includes a series of able- bodied walking experiments while wearing an exoskeleton while experiencing a slip or trip on an instrumented

treadmill. A Fall Classification Algorithm (FCA) will predict the occurrence and classify the type of a fall using center of mass kinematics measured from the experiments. Specific Aim 2 focuses on identifying the exoskeletal configurations that minimize the magnitudes and locations of impact forces on the user and

applying this information to design an Impact Mitigation Controller (IMC) that reconfigures the exoskeleton’s joint angles into these safer configurations. Optimizations of a 3D musculoskeletal model developed in the OpenSim Modeling Suite will minimize the impact forces at the head, joints (neck, shoulder, elbow, wrist, hip,

knee, and ankle), and long bones. These experiments will be repeated on an instrumented crash-test dummy and the potential for injury of the reconfigured positions will be statistically compared to the baselines. The trained FCA and IMC will be validated in real-time in a repeated set of experiments with able-bodied

participants. All studies will be performed at the Louis Stokes Cleveland VA Medical Center in facilities associated with Advanced Platform Technology Center, such as the Motion Study Laboratory directed by Dr. Triolo. Dr. Hnat has identified multiple training aims within her proposal, including developing musculoskeletal models and

performing optimizations with them, designing experiments that represent realistic falling scenarios, learning and applying statistical analyses to collected data, practicing hardware design and implementation, and understanding the challenges experienced by individuals with SCI. The mentoring team she has assembled is

well equipped to help the candidate achieve the training aims outlined within this proposal as well as transition to a successful independent researcher in rehabilitative technologies.