File(s) under embargo
10
month(s)9
day(s)until file(s) become available
Access and Inclusion in the Development of Somatosensory Neuroprostheses
For millions of people around the world living with amputations, somatosensory neuroprostheses (prosthetic devices that provide sensory feedback from the missing limb) could offer major functional improvements and restore a sense of independence in their daily lives. Despite these potential benefits, neuroprosthesis research studies struggle with low recruitment, especially among people from marginalized backgrounds. The burden of amputation itself is unequally distributed along axes including race, geography, and income level, with Black, rural, and low-income communities seeing a higher than average incidence of amputation. These same communities are often excluded from research and novel technologies due to a combination of research hesitancy and low access to research opportunities and novel devices. The goals of this dissertation are two-fold. First, we seek to illuminate the perspectives of people with amputations regarding medical research, research participation, and somatosensory neuroprostheses in order to better serve their needs. Second, we seek to employ computational modeling techniques to support the development of viable somatosensory neuroprostheses that leverage existing, approved clinical devices and may therefore be easier for future users to accept and adopt.
The first aim of this dissertation lays out a human-centered design (HCD) framework that we use to illuminate prosthesis users’ perspectives and co-create solutions and recommendations with them. This process yielded insights into the access concerns and physical invasiveness concerns that are barriers to research participation and neurotechnology uptake and provided recommendations for the neurotechnology field to more closely engage with prospective users. Encouraged by some of these insights, namely participants’ higher comfort level with established technologies compared with experimental ones, we then examine two potential neuroprosthesis approaches that leverage existing clinical technologies: dorsal root ganglion stimulation (DRGS) and spinal cord stimulation (SCS). In the second aim, a computational model of DRGS reveals that the specific neuronal morphology and microanatomy of the DRG makes it possible for epineural electrodes of the type that are already in clinical use for chronic pain management to achieve selective activation of neural fibers in the DRG, suggesting that this approach could enable the elicitation of focal, targeted percepts in the missing limb. Finally, the third aim describes a patient-specific computational model of SCS in the spinal cord of an individual participant that can provide the basis for a testbed for identifying stimulation locations and parameters that can evoke desired percepts. Such a testbed can reduce the need for extensive testing in participants, and for a future clinical device it can reduce the time patients need to spend in the clinic to calibrate their device.
This dissertation marks a meaningful step toward accessible and inclusive neuroprosthesis development through both demonstrating how clinical technologies and computational modeling can be used to make neuroprostheses viable and adoptable, and laying out recommendations for the field based on the concerns and priorities of prospective users.
History
Date
2024-09-01Degree Type
- Dissertation
Department
- Neuroscience Institute
Degree Name
- Doctor of Philosophy (PhD)