Following spinal cord damage from trauma or disease, skeletal muscles distal to the point of damage become paralyzed due to disrupted neural conduction. In high-level spinal cord injury/damage, there is a great need for a method that can substitute for the voluntary control in order to regain self-mobility, environmental control, and computer access. Each year about 15,000 spinal cord injuries occur in the US. Majority of these cases survive and depend on others for their basic needs. The average life expectancy of this population is 40 years. Current Solution: 'Brain-Computer Interfaces'have been developed to extract the volitional control information from various brain cortices. Activity of single neurons is recorded with micro electrode/wire arrays implanted and interpreted with advanced signal processing algorithms. Shortcomings: There remain two major problems after many years of research that are inherent to single spike recording method from the cerebral cortex. First, the population of neurons recorded from changes day to day, thus requiring a training session for the signal processing algorithm before each use. The number of good electrodes that record neural activity in an array (yield) is low and the single spike signals are lost completely after sometime due to glial cell growth around the electrode. Second, the information provided by each neuron is very noisy. A very large number of neurons need to be sampled to achieve stable and finely tunable command signals. This requires many electrode arrays implanted in multiple brain areas. Our Proposal: The alternative method proposed here is to extract the volitional motor signals from the proximal spinal cord that is still intact above the site of injury and use the population activity of the axons in the motor tracts rather than single spikes. The distal portions of the severed motor axons go through Wallerian degeneration. However, the proximal part of the axon continues to function years after injury since its connection to the cell body in the brain remains intact. Spinal cord approach has at least two important advantages. First, the recorded neural signals will be strongly coupled to the motor function due to closeness of the spinal cord to the motor apparatus in the signal path. Second, the neural recordings will be much more stable because the method relies on the population activity rather than single spikes.

Public Health Relevance

Each year about 15,000 spinal cord injuries occur in the US. Majority of these cases survive and need help for their basic needs. The average life expectancy of this population is 40 years. This project aims to develop an interface between the subject's brain (via the spinal cord) and a computer so that the paralyzed individual can control his/her own wheelchair and other equipment in the house without needing help from a caregiver. A device like the one proposed here that can provide them with environmental control (such as lights, room temp, TV, etc.) or computer access is priceless to the individual. Also, the potential contribution of this project to the economy is in the range of billions of dollars due to savings of health care cost by providing these paralyzed individuals with means to live more independent lives.

Agency
National Institute of Health (NIH)
Institute
National Institute of Neurological Disorders and Stroke (NINDS)
Type
Research Project (R01)
Project #
5R01NS072385-02
Application #
8328925
Study Section
Neurotechnology Study Section (NT)
Program Officer
Ludwig, Kip A
Project Start
2011-09-15
Project End
2015-08-31
Budget Start
2012-09-01
Budget End
2013-08-31
Support Year
2
Fiscal Year
2012
Total Cost
$283,637
Indirect Cost
$86,762
Name
Rutgers University
Department
Biomedical Engineering
Type
Schools of Engineering
DUNS #
075162990
City
Newark
State
NJ
Country
United States
Zip Code
07102
Guo, Yi; Gok, Sinan; Sahin, Mesut (2018) Convolutional Networks Outperform Linear Decoders in Predicting EMG From Spinal Cord Signals. Front Neurosci 12:689
Wang, Lu; Freedman, David; Sahin, Mesut et al. (2016) Active C4 Electrodes for Local Field Potential Recording Applications. Sensors (Basel) 16:198
Gok, Sinan; Sahin, Mesut (2016) Prediction of forelimb muscle EMGs from the corticospinal signals in rats. Conf Proc IEEE Eng Med Biol Soc 2016:2780-2783
Guo, Yi; Foulds, Richard A; Adamovich, Sergei V et al. (2014) Encoding of forelimb forces by corticospinal tract activity in the rat. Front Neurosci 8:62
Guo, Yi; Mesut, Sahin; Foulds, Richard A et al. (2013) Corticospinal signals recorded with MEAs can predict the volitional forearm forces in rats. Conf Proc IEEE Eng Med Biol Soc 2013:1984-7
Hua Meng; Sahin, Mesut (2013) An electroacoustic recording device for wireless sensing of neural signals. Conf Proc IEEE Eng Med Biol Soc 2013:3086-8
Prasad, Abhishek; Sahin, Mesut (2012) Can motor volition be extracted from the spinal cord? J Neuroeng Rehabil 9:41
Prasad, Abhishek; Sahin, Mesut (2011) Chronic recordings from the rat spinal cord descending tracts with microwires. Conf Proc IEEE Eng Med Biol Soc 2011:2993-6