Workshop Report

Rehab Home
Falk Center

Part 1: Background | Part 2: Summary

Report of the

Falk Workshop on Innovations in Neurorehabilitation:

Future Possibilities for Technology-Assisted

Neuromotor Assessment and Movement Therapy

November 2-3, 2001, Milwaukee WI

Abstract

Part I: Background

Prior to the Workshop, expert consultants were asked to complete a survey, which included providing bullets that addressed overriding clinical needs, barriers, gaps in knowledge, and emerging technologies in neurorehabilitation. These were collected and placed on the web as a resource.

The first day (November 2) started with opening presentations by representatives from the key teams that where present: MU-MCW (by Dr. Harris); UW-Madison (by Dr. Bach-y-Rita); MIT (by Dr. Hogan); RIC/NWU (coordinated by Dr. Rymer); VA/Stanford (by Drs. Burger, Lum and Van der Loos); and the Telerehab RERC (by Dr. Winters).

Participants were then divided into two breakout sessions of roughly 20 each to address key needs, barriers and gaps in knowledge. Here is a brief synopsis of these initial discussions.

Lower Extremity Group (targeting mobility and posture)

Moderators: Drs. Gerald Harris and Brian Schmit

Clinical Needs and Barriers

Barrier: Reimbursement -- can effective tools, etc., reduce this barrier?

Need: More effective collaboration between clinicians & researchers

Need: Better outcomes measures & reality based measures

Need: Quantitative information -- better functional assessment tools

Need: Automated data handling -- especially for monitoring of function

Gaps in Knowledge

Biological linkages between: Disease/Impairment/Function Limitation/Disability

Objective measures of impairment / function, and objective outcomes

Modeling and effects: Relate pathophysiology to impairment in function

Linkage between incremental scientific knowledge and functional changes.

Acknowledge our limitations to incrementally solve the problem, and perhaps the need for a bioinformatics center that can integrate clinical and incremental technical data/modeling

Recognition of sensory / visual and cognitive deficits and their impact (not just motor deficits)

Technical gaps in knowledge (actuator and sensor technologies, engineering methods)

Emerging Technologies:

Virtual reality to incorporate visualization with locomotion and other therapies

Robotics technologies for locomotion and posture

Applications of Imaging: neural imaging (e.g., functional MRI)

Sensor-based technologies (e.g., biochemical monitoring (e.g., instantaneous O2 consumption), biomechanical monitoring, real time sensing)

Ties to pharmacological interventions (e.g., intrathecal baclofen/BOTOX for spasticity)

Telerehabilitation, as related to ties to neurorehabilitation needs

Computer science advances in multi-degree-of-freedom dynamic control

Upper Extremity Group (targeting reaching and manipulation)

Moderators: Drs. Jack Winters and Robert Scheidt

Clinical Needs and Barriers:

Need: For consumer-driven technology development

End user (e.g., client with disability, practitioner) must be integral part of research team.

Special need: evaluate assessment and therapy in the home environment (greater ecological validity), e.g. through telerehabilitation tools.

Need: To focus on functional improvement, not just improving "numbers"

Technology must focus on patient's goals (e.g., to walk).

Numbers are dependent on the measure used, and do not necessarily capture what therapists observe. There is potential for misuse of numbers.

Clinicians may hesitant to use numbers, and may especially not want new numbers.

Need: For technology that enhances efficiency and efficacy, including through approaches that are profoundly novel, and break hard barriers.

Clinicians need to help guide the engineers developing tech-based therapies, e.g. identifying goals, therapeutic approaches, and what outcomes measures are most important

Proving therapeutic efficacy of devices is a challenge.

Practical Barriers:

The human touch is very important

Mind-set against use of technology perceived as simply engineering toys.

Robotic therapy may be less well accepted in some settings.

Complexity of some technological tools. (Einstein's rule: Keep as simple as is possible, but no simpler)

Therapists productivity-driven; not often trained in technology tools.

Gaps in Knowledge:

Overriding need for scientific evidence for current treatments

Dynamics of neuro-recovery and motor learning.

What are optimal therapy techniques, dosage patterns?

What should we train (e.g., forces, kinematics, synergies, ADLs)?

Knowledge of behavior of complex neuromusculoskeletal systems.

The morning of Day 2 started with a collection of invited position statements by experts, related to the primary theme for all subsequent discussions: Future directions and opportunities for technology-assisted neurorehabilitation assessment and movement therapy.

In preparing for the breakout sessions targeting this issue, participants were told to consider that products enter the market not only by the classic technology-push and clinical-pull approaches, but now also by healthcare consumer-pull. During this second day and through subsequent interactions, an valiant attempt was made to distill the many discussions and comments into a finite collection of key statements, in the form of finite set of recommendations, that could represent the collective sense of the assembled participants.

Part II: Summary

These top 7 overriding statements, roughly ranked, represent the primary outcome of the Workshop, and may be of interest to a range of stakeholders (e.g., researchers, clinicians, consumers, funding agencies). Each consists of a core statement, in most cases followed by a set of bullets.

Participants felt that technology-assisted rehabilitation tools need to be optimized to address needs of specific patients, but must do so as part of an integrated approach that includes other interventions. Such optimal strategies remain illusive. There are a number of pressing scientific and clinical issues to be addressed, most related to gaps in knowledge where utilization of technology-assisted rehabilitation tools may help promote a paradigm shift in rehabilitative care:

Timing and Intensity. There is currently a major, overriding gap in knowledge regarding optimum intervention strategies. Recent evidence is forcing a re-evaluation of the optimum timing, intensity and duration of intervention. For instance, workshop participants noted that there is evidence that intensive therapy of "chronic" individuals with stroke who are beyond the conventional continuum of care for services can show strong functional benefit (e.g., results with constraint-induced movement therapy). The value of long-term exposure must be re-evaluated.

Components of Therapy. Participants felt that there is also limited evidence regarding the optimum components that should constitute an intervention program. This menu of possibilities is large, and includes

unilateral and bimanual modes,

actively resistive,

passive movement,

actively assisted movement,

consideration of accuracy of movement in relation to position or forced based systems, etc.,

degrees-of-freedom of the device and required therapeutic interventions,

sequence of training (e.g., between proximal arm vs hand),

importance of sensory information in facilitating functional rehabilitation,

open-chain vs. closed-chain and concentric vs eccentric control,

EMG-induced stimulation therapy.

Context/Infrastructure of Healthcare Delivery System. Innovation through technology-assisted neurorehabilitation must recognize constrains associated with the existing rehabilitative healthcare infrastructure, and in most cases work within this system.

Optimizing Delivery Based on Initial and Ongoing Assessment. This implies the need to better understand/quantify the impact of the extent and location of lesion on response to therapy.

Several participants noted that there is an evolving clinical horizon for the application of incremental quantitative feedback in the application of functional therapy.

Participants felt that there is a need for low-cost rehabilitation tools that can extend consumer access to supervised therapy into the home. There were a number of related observations:

There are opportunities for automated data collection for monitoring/telemonitoring neurorehabilitation interventions (e.g., home activity assessment and motion analysis, electronic health records integrated with telehealth).

This could possibly build on consumer-oriented, community-based educational model for relearning life skills following neural injury (e.g., "distance learning" could employ web-based and telehealth tools).

Approaches could capitalize on emerging information technology infrastructure, such as Web pages and mobile devices, to facilitate life-skills problem solving.

To evaluate therapy in the home environment, consideration of psychosocial factors needs to be integrated in to the approach.

Such infrastructure can benefit both consumers and therapists, providing opportunities for large-scale universal access to larger populations, as well as promoting standardization and research opportunities related to large scale-up in population sizes.

However, there is a gap between elements of home exercise program as currently prescribed and the availability of current off-the-shelf instrumentation and software. This must be addressed. This represents both a concern and an research opportunity.

Most participants felt that there was great potential for rehab robotics. Specifically, there are opportunities for technological tools such as robots, especially machines that are are flexible in their integration into clinical and research practice.

Participants within the upper extremity group expressed the following:

Such mechanical devices can evaluate, treat and document improvements in the movement components that are strategic barriers to functional tasks.

Robotic interfaces can give us great insight into movement barriers (e.g., specific limitations in strength, coordination patterns)

The information that we can gain from the robot may provide valuable understanding into the physiology underlying impairment and recovery and/or enhancement of motor function.

And yet, the need for and efficacy of actuated robotic rehabilitation tools must be further demonstrated, with the onus on the advocate:

Such machines must be safe - this is essential.

Actuated machines should demonstrate value that is over and above mechanically passive devices that measure and/or assist movements that may be more appropriate for subsets of tasks for certain subject populations.

Another observation was the potential for melding therapeutic robotics with assistive robotics, and use of therapy robots in a functional environment. For instance, a smart controller that adaptively adjusts the amount of mechanical assistance that is given so that the patient is always encouraged to use his/her own effort as much as possible.

Participants in the lower extremity group had many of the same comments as listed above. Specifically, they also felt that there are considerable opportunities for the application of robotic-assisted locomotor training in a multitude of rehabilitative areas.

Participants felt that there are many opportunities for technology-assisted quantitative measurement of specific pathophysiologic neural and mechanical parameters in the context of functional activities.

This includes active versus passive metrics.

This includes new quantitative measures of sensory and motor function (e.g., evolution of new biochemical and mechanical sensors, emerging dynamic imaging technologies).

Most participants felt that there are unique opportunities for the application of virtual reality in the assessment and therapy of neurorehabilitation. The following observations were made:

An advantage is the provision of a simulated environment is that allows strict control of strategic variables within the virtual environment.

Another advantage is that it may add safety to therapy.

It also allows sensors to be used in a small space.

It expands biofeedback principles and offers enhanced motivational opportunities.

However, some participants noted that this is not a new concept and to date has had limited success, and the onus is on the advocate to make the case as to why therapy in a virtual world is better that therapy in the real world (in which the patient must ultimately function).

Participants suggested that there was considerable need (and thus an opportunity) for planned, structured research coordination that address the integration of basic physiology, neuromuscular metrics, and functional needs, and can serve to help bridge basic science and clinical practice. This suggests:

A multi-center approach to data acquisition.

Applications in bioinformatics, and approaches such as data mining, intelligent agents and genetic algorithms/networks for making sense of large diverse collections of data.

Some participants noted that perhaps those assembled for this Workshop could help coordinate such as effort, targeted toward technology-assisted neurorehabiliation.

Many participants observed that there are opportunities for ties between neurorehabilitative bioengineering and emerging areas of bioscience. Two stood out:

Cellular, molecular and tissue engineering (including stem cell research) may provide rewarding collaborative opportunities to apply technology to facilitate functional recovery following neural injury.

Technology-assisted movements may be ideal for providing the type of repetitive, carefully planned interventions that could help train implanted stem cells to learn their newly desired function..

Novel pharmacological agents used in neurorehabilitation may be more systematically integrated with other methods such as functional assessment and therapeutic interventions. Examples include:

Integration of new sensor technologies and quantitative measurements to study the temporal effects of drug delivery on movement (e.g., to determine an optimum dosage that maximally enhances or minimally impairs performance).

Use of pharmacologic agents to examine subtle mechanisms of impairment (e.g., that influence a type of receptor site).

Optimum therapeutic intervention strategies that combine dosage for pharmacologic therapy (e.g., for spasticity) and movement therapy (e.g., timing, intensity).

___________________________________________