Critical Review of the Available Utilization Protocols: How Can Exoskeletons Be Used?
As stated throughout this chapter, exoskeletons designed for people with disabilities are currently used for rehabilitation and health purposes only and typically only used in clinical settings. However, we review the target populations for the existing devices, their requirements for use, and exclusion criteria that may limit their use by specific people. We also look forward and address exoskeleton use as AT devices, discussing how exoskeletons may be used and which applications and functional tasks may be possible.
Lower Limb Exoskeletons
Most LLE exoskeleton companies list persons with a SCI as their target population for use. The identified populations for each technology are listed in Table 6.1. Of the LLEs, the FDA-approved systems are only intended for use by users with paraplegia and only specific levels of SCI at that: “ReWalk is for people with paraplegia due to spinal cord injuries at levels T7 (seventh thoracic vertebra) to L5 (fifth lumbar vertebra) when accompanied by a specially trained caregiver. It is also for people with spinal cord injuries at levels T4 (fourth thoracic vertebra) to T6 (sixth thoracic vertebra) where the device is limited to use in rehabilitation institutions” (U.S. Food and Drug Administration 2014, fifth paragraph). Persons with other levels of injury, including incomplete tetraplegia with significant remaining arm function, are also using LLEs in clinical and research studies. Uniquely, the Rex LLE can be used by persons with SCI up to C4/5, similar to those who can use a joystick-driven power wheelchair. SCI is the largest documented etiology of mobility impairment studied during LLE use, as well described in the Federici et al. 2015 review.
Individuals with stroke make up the second-most-targeted disease population for LLE studies (Federici et al. 2015). In particular, the variable assist feature in the Ekso GT, and similar control methods, are claimed useful for stroke rehabilitation, as well as for incomplete SCI. In the broadest sense, Ekso claims that any person with paralysis or hemiparesis due to neurological disease or injury may be appropriate for use of the Ekso GT. This statement mirrors the literature as well, as after SCI and stroke, “gait disorders of unspecified etiology” were the only other category of disability identified by Federici et al. (2015). In terms of potential use as an AT, and ignoring the technical limitations of current exoskeletons, it is reasonable to assume that any mobility-impaired population currently part of investigational studies would be an appropriate candidate for using exoskeletons as AT mobility devices.
Regardless of the target population for exoskeleton use, there are several general inclusion and exclusion criteria for use of LLEs. Federici et al. (2015) provided some information about these criteria, with more found on inspection of companies’ websites. Users must be medically cleared by a physician, with consideration of bone density, contractures, spasticity, orthostatic hypotension, cognitive function, and arm function sufficient for walker or crutch use.
Exoskeleton technologies can be controlled in several different ways. Yan et al. (2015), in an in-depth review about control strategies of broadly defined LLEs, characterized seven different strategies. For our purposes, only two strategies are employed by the devices reviewed here: predefined joint trajectories and cooperative or model-based control. The predefined joint trajectory strategy incorporates well-known patterns of healthy gait, scaled appropriately to a user’s size and the desired gait parameters (e.g., walking speed and stride length). This strategy is common, forming the basis of at least one mode of control on all identified LLEs described previously. It is appropriate for users with complete paralysis and thus dependent on full assistance from the exoskeleton.
The cooperative or model-based control strategy uses an integrated human-device model and typically combines joint trajectory control with a human-based control signal, which is a signal somehow derived from or applied to the user directly. The Indego LLE combines actuator-controlled joint trajectories with an FES controller to effect cooperative control of external and human power (Ha, Murray, and Goldfarb 2016). The HAL LLE has been demonstrated with cooperative control using residual EMG signals of the user (Cruciger et al. 2014). The Ekso GT LLE has its variable- assist operation modes, whereby in “Adaptive Assist” mode the users’ own muscle activity participates in the joint trajectory control strategy, with the external actuators compensating as needed.
What are the limitations currently on use of exoskeletons in the community? Most LLEs are not approved for use in the United States (with the exception of ReWalk and very recently Indego in March 2016), but most do have approval in Europe for use at home. In these cases, the described operation modes of the previous sections apply. For LLEs in the United States, it is interesting to note that the FDA classified these as class II medical devices and placed “special controls” on the use of the device. This not only includes their use only with a companion, but also specifies the training necessary for the user and companion and restricted environments of use (e.g., small slopes and no stairs).
Adapting (Cowan et al. 2012), we can consider a reduced scope of mobility-based functions when considering new technologies, including postural control; balance; transfers (or donning and doffing the exoskeleton); walking on level ground and turning; stair climbing; walking on slopes (both up and down and across slopes) and uneven terrain; transitions between sitting, standing, and walking; and using transportation. Little information about any of these functions, besides basic walking distance and speed, is provided in the literature.
Bryce, Dijkers, and Kozlowski (2015) have recently presented a framework for assessing LLE usability. This was not done solely with AT use in mind; rather, it considered a wide array of uses for LLEs, including early and late rehabilitation, exercise, and as a wheelchair replacement. The proposed framework documented six components to consider when assessing LLE use: functional applications, personal factors, device factors, external factors, activities, and health outcomes (Bryce, Dijkers, and Kozlowski 2015). This framework may prove useful when considering future LLE use for mobility, both at home and in the community. Measuring and documenting exoskeleton features and predicted functional impact may guide device prescription, as well as provide insight to guide future device development.
Depending on the environment of use and activities performed, there would be different requirements to allow an LLE exoskeleton to function effectively as an AT. Some of these are summarized here. Because many activities require the use of the upper extremities (e.g., cleaning, cooking, shopping), it would be important to be able to have the users’ hands free while standing for short periods of time. Battery life and charging would also be important considerations. Most users would likely require a minimum battery life to make donning and doffing the device worthwhile. Potentially, inductive charging pads might be one option so that the device would not need to be plugged in for charging. Currently, many devices have restrictions about the minimum seat dimensions and height from which they can rise. If used in a specific setting, chairs that met these criteria could be used. If used in other settings, then a way to appraise the appropriateness of different seating surfaces or more flexibility in types of seating that can be used would be beneficial. Ease of doffing and donning is also an important consideration, especially in terms of toileting and dressing. Currently, level changes (e.g., curbs, slopes, and stairs) are challenging with most devices. This is an area for development. For outdoor mobility, walking speed would need to be sufficient for street crossing within timed light cycles. In the future, it may be possible to design exoskeletons to enable people to perform complex lower limb activities, such as car driving (including transfers in and out and use of pedals) or sports. Finally, safety is perhaps the most significant requirement moving forward, a topic more fully addressed in the section on future directions.