Wearable and Wear-nothing Approaches for Assistive Technology Assessment

Ed Lemaire, PhD; Professor, Physical Medicine and Rehabilitation, University of Ottawa; Affiliate Investigator, Clinical Epidemiology Program, Ottawa Hospital Research Institute

Wearable technology has evolved to provide accessible and cost effective opportunities to measure human movement and generate quantitative information to assist clinical decision-making. This presentation will examine a variety of technologies and outcome analysis approaches that can provide useful and immediate information to the healthcare provider. Examples include inertial measurement units and machine learning to evaluate movement quality and fall risk, augmented reality technology for realtime posture analysis, and markerless tracking for moving beyond discrete biomechanical analysis to continuous movement assessment. By combining these technologies, assistive technology providers can work in a new era where data processing is replaced by real-time reporting and immediate evidence-based practice is in their hands.


Personalized rehabilitation: Mapping the links between prosthetic device parameters, motor capacity, and clinical outcomes for optimizing interventions

Matthew J. Major, PhD; Assistant Professor, Northwestern University, Feinberg School of Medicine; Research Health Scientist, Jesse Brown VA Medical Center

The selection of prosthetic device parameters to restore function of persons with lower-limb loss can be framed as an optimization problem to maximize a given performance outcome. The choice of particular outcomes, such as locomotor stability, comfort, or metabolic cost, is a shared decision amongst the rehabilitation team and patient, and depends on a patient’s functional capacity and motivation. Historically, the selection of prosthesis parameters to match components with patients has been constrained to those fixed mechanical properties of commercial devices. Classical comparative studies have suggested an effect of commercial prosthesis design on clinically-relevant outcomes, but these studies lack reliability and the results are often inconsistent. However, recent parametric studies have begun to systematically probe the influence of prosthesis parameters (i.e., mechanical function) on performance outcomes through the use of novel methodologies and experimental prosthetic technology. Results from these parametric studies generate maps defining the relationships between select parameters (e.g., stiffness, damping, roll-over geometry) and a desired outcome. Moreover, these maps can be expanded into a multidimensional landscape when considering factors of patient motor capacity (e.g., muscle strength, sensory feedback) as an additional covariate. Through these correlate maps, parameter values can be identified which maximize a given performance outcome, thereby providing an objective framework to optimize prosthesis designs. Importantly, by accounting for person-specific variables pertaining to motor performance, this framework can yield predictions of performance outcomes for individual patients. This review will discuss: 1) a selection of novel experimental techniques being implemented to populate these correlate maps, and 2) how an iterative optimization approach can deliver personalized lower-limb loss rehabilitation interventions when integrating targeted physical therapies that encourage self-organization with the prosthesis.


Development and Dissemination of the Northwestern University Flexible Sub-Ischial Socket Technique

Stefania Fatone, PhD, BPO(Hons); Professor, Northwestern University, Feinberg School of Medicine

Current standard-of-care Ischial Containment sockets for transfemoral amputees fit intimately with the ischium, limiting hip motion and contributing to socket discomfort. Hence, we developed a technique to fabricate and fit a sub-ischial socket aimed at improving socket comfort without compromising function. Both the Northwestern University Flexible Sub-Ischial Vacuum (NU-FlexSIV) Socket and the Northwestern University Flexible Sub-Ischial Suction (NU-FlexSIS) Socket have lower proximal trim lines that do not impinge on the pelvis and include compliant materials in their fabrication to improve user comfort. Since the socket we developed is a technique and not a product, dissemination has relied on being able to successfully communicate and teach the technique to other prosthetists. To facilitate teaching of our technique we developed a simplified approach to casting, rectifying and fitting the sub-ischial socket supported by documentation in the form of a socket work form, mold reduction algorithm and rectification mapping. This presentation will describe development of our socket technique and our experiences disseminating the technique to prosthetists via mechanisms that include hands on workshops, online webinars, conference presentations and peer-reviewed publications. We anticipated that dissemination of the sub-ischial socket technique might be challenging given that removal of the proximal brim challenges conventional understanding of the biomechanics of transfemoral sockets wherein “locking onto the pelvis” is believed to stabilize the socket in the coronal plane. While we encountered some initial skepticism regarding the potential success of the technique it did not last long. Since launching the technique in 2015, we have experienced ongoing demand nationally and internationally for hands-on workshops and invitations to present at conferences with increasing anecdotal evidence of clinical implementation. To date over 200 prosthetists have taken our hands-on workshop, we’ve published 8 peer-reviewed articles and given nearly 60 presentations at conferences, including 18 invited and keynote lectures on the development and research related to this sub-ischial socket technique.


Using biofeedback systems to improve gait of lower-limb prosthetic users

Dr. Jan Andrysek, PhD, PEng; Senior Scientist, Bloorview Research Institute, Holland Bloorview Kids Rehabilitation Hospital

Biofeedback, the practice of providing direct, real-time biological information, can aid the rehabilitation of individuals with musculoskeletal disorders including amputations, and accelerate and optimize recovery of aspects of physical function and movement. A major benefit of biofeedback systems is the ability to provide feedback to reinforce personalized physiotherapy goals and good walking pattern habits. In this talk, I will provide an overview of the state of biofeedback systems within rehabilitation of lower-limb prosthesis users, and present our original research aimed at advancing the use of wearable haptic-based biofeedback systems for gait training.


The Impact of Cognition of Prosthetic Outcomes

Dr. Michael Payne MSc, MD, FRCP(C); Associate Professor, Western University

Most major lower extremity amputations are performed due to one or both of diabetes and peripheral vascular disease. These dysvascular medical conditions are frequently (56% of people admitted to a rehabilitation program) associated with some degree of cognitive impairment. Typically, these cognitive impairments present as deficits in memory, attention, and visuospatial perception, all of which are important for achieving success with prosthetic rehabilitation.

Cognitive impairment related to traumatic brain injury associated with traumatic upper extremity amputation has not been studied extensively, although likely diminishes outcomes in rehabilitation-intensive protocols such as following targeted muscle reinnervation.

Both cognitive and prosthetic performance have various clinical measures that can be useful to evaluate for a given patient. Dual-task testing is a method of measuring performance of two independent tasks, performed concurrently, that may challenge the cognitive capacity of the individual. When the cognitive capacity is exceeded, performance may suffer in one or both tasks. Among people with amputations, both gait and cognitive responses deteriorate when performed together, although gait appears to be more commonly prioritized over the cognitive task.

Ambulation with a prosthesis is a cognitively demanding task for many patients, especially in pursuing the goal of reintegration to the community. The dual-task costs of cognitive and gait tasks do not change over time, suggesting that cognitive rehabilitation is an important component of prosthetic training in addition to the physical use of a prosthesis.


On the Design and Control of Partial-Assist Powered Orthoses for Broad Patient Populations

Dr. Robert Gregg, PhD; Eugene McDermott Professor, University of Texas at Dallas

The majority of assistive exoskeletons (i.e., powered orthoses) are designed to rigidly track time-based kinematic patterns using highly geared actuators, which prevents users from moving their joints freely without help from the exoskeleton. This class of exoskeletons is appropriate for individuals with spinal cord injury, but many patient populations present with weakened volitional control of their lower extremities (e.g., stroke or musculoskeletal disorders) and only require partial assistance to regain mobility. This clinical need motivates novel design and control methods for powered orthoses that are more compatible with human interaction. In order to assist or augment volitional human motion, orthotic joints must be backdrivable and the control strategy must be invariant to the user’s activity. This talk will present the design philosophy behind two generations of backdrivable powered orthoses, which utilize torque-dense motors with low-ratio transmissions. A trajectory-free control framework shapes the kinetic and potential energies of the human body rather than prescribing joint kinematics. Preliminary human subject experiments demonstrate the user-cooperative and task-invariant nature of this design and control approach. The talk will conclude with a discussion of future work including connections to powered prostheses for agile amputee locomotion.