The NIH neuroprosthesis program has fostered so much success in the area of cortically controlled neuroprostheses that the FDA has approved multiple human trials to test the safety and efficacy of cortical implants for brain machine interfaces (BMI). One important application of BMI technologies is the direct cortical control of prosthetic limbs. Recent advances in this field have led to the creation of the most capable prosthetic arms yet developed, including the DEKA 'Luke arm'and Johns Hopkins APL 'Modular Prosthetic Limb'. However, a critical gap in this effort is the lack of somatosensory feedback which is needed to support propriception and tactile sensations for the artificial limb. Without these sensations, users will never achieve maximum benefit from these advanced limbs, because without sensory feedback, these devices will remain as numb, extracorporeal 'tools', rather than integrated fully functional limbs. Our goals are twofold: to better understand the nature of sensory feedback and the way in which peripheral sensory activity is conveyed to primary somatosensory cortex (S1), and to develop a somatosensory neural interface (SSNI) that will provide the user with proprioceptive feedback for their neuroprosthesics limb. We have previously proposed that primary afferent microstimulation (PAMS) in the dorsal root ganglia (DRG) can be used to deliver surrogate somatosensory feedback to the central nervous system. We have demonstrated that in cats, PAMS can recruit small populations of afferents from a variety of sensory modalities (Gaunt et al. 2009) and that this stimulation can transmit meaningful activity to S1 (Weber et al. 2011). The success achieved during the development of this animal model generated a number of new questions and hypothesis upon which a series of new experiments are proposed. Specifically, these experiments focus on characterizing the ability of PAMS to 1) transmit sensory information to S1 in anesthetized cats when the PAMS patterns are based on neural activity recorded in the DRG during movement, 2) transmit discriminable sensory information to S1 in anesthetized cats when the PAMS patterns are based on fabricated static and dynamic inputs, and 3) transmit discriminable sensory information to S1 in awake standing cats, useful for modifying postural responses to ground support perturbations. These experiments range from further investigations of the capabilities of PAMS to testing the ability of PAMS to predictably modify motor behaviors. This work will further the development of a SSNI, critical for the future of BMI based prosthetic limbs, as well as address fundamental questions regarding the role of sensory feedback in the control of normal motor behaviors.
Trauma, vascular disease, and diabetes are leading causes of limb amputation and sensory deficits in both civilian and military populations. The number of people living with the loss of a limb is expected to more than double to 3.6 million by 2050. Thus, there is an urgent and rapidly growing need for advanced prosthetic limbs that can restore the motor and sensory functions that are lost after amputation. The goal of this proposal is to develop a technique for providing proprioceptive sensations to users of prosthetic limbs using patterned electrical stimulation of sensory neurons in the dorsal root ganglia (DRG). This approach is similar, in principle, to that of the cochlear implant which uses patterned electrical stimulation of auditory nerves to restore hearing to people with profound deafness. The proposed study will determine if DRG stimulation is effective in: 1) delivering proprioceptive information to the brain and 2) providing feedback that is useful for maintaining balance during postural perturbations.
|Boninger, Michael L; Wechsler, Lawrence R; Stein, Joel (2014) Robotics, stem cells, and brain-computer interfaces in rehabilitation and recovery from stroke: updates and advances. Am J Phys Med Rehabil 93:S145-54|