Present Developments and Applications of Neural Interfaces
Prostheses have been in use since the ancient Greeks, but the technology of today has only enhanced the capabilities of prosthetic limbs. The newest innovations in prostheses involve the use of neural interfaces. Neural interfaces are devices used to convert chemical signals sent from the brain through neurons to electrical signals that can be manipulated by technology or vice versa. They can be thought of as a translator between two vastly different languages except the parties talking are a computer and a person’s brain. Currently, research has been working to find more methods that more effectively translate signals in one direction or the other. It has also focused on using existing methods of translating and sending signals for the rehabilitation of the neural function of patients rather than just the replacement of lost function.
To understand how neural interfaces work, it’s important to know what the actual interfaces do to translate between the nervous system and current technology. To do this we turn to our typical neural interface, which was the array of cells, sensitive to the changes in electrical activity that occur when neurons depolarize. The example being discussed is an interface with an array of 256 cells, developed by Aziz et al. Neural interfaces sense the change in neural activity by comparing the electrical activity of the cells at two different instances; these signals are filtered because the frequencies of action potentials are generally from .1 Hz to 10 kHz. To compare these two instances in time, it is necessary to hold the first measured voltage within the circuit while the second voltage is being sampled accomplished by the Correlated Doubling Sampling Circuit. The circuit then compares the two signals by taking a voltage difference, which we can associate with a specific change in time based on the sampling rate (Aziz et al. 2005). If viewed in general, we can take the differences calculated by each cell’s circuitry and see overall changes in the area covered by the array of cells. These spatio-temporal patterns of neural activity over an area can then be correlated to different applications such as being associated with actions to be carried out by prosthesis.
When trying to develop an idea of what different patterns of changes in neural activity mean, it makes sense that having a higher definition image makes differentiating different intended signals easier. We could liken it to focus in vision by imaging a silhouette of an image on a screen. If the silhouette cast on the screen is blurry and faded at the edges it is difficult to make out what the object is, but if we can sharpen the lines and see more clearly the edges of the shadow we are much more likely to correctly guess what we are looking at. It was with this idea in mind that the use of carbon nanofibers as electrodes, rather than older methods such as the arrays of cells, makes sense. It is evident from the name that carbon nanofibers exist on a much smaller scale than the large arrays of cells. Yu et al used a novel idea of taking carbon nanofibers and placing those in a 3 dimensional microelectrode array called Vertically Aligned Nanofibers (VANF). Yu et al found that the soma (body) of the neurons absorbed the tips of the VANFs through endocytosis (the process by which the plasma membrane of a cell engulfs and brings in large molecules) allowing them to record intracellular neural activity. Because the VANFs penetrate the cell, they are also capable of directly recording the release of chemical neurotransmitters (form by which neurons communicate with each other at synapses). Through the simultaneous recording of both the electrophysiological and neurochemical forms of communication, images and changes to neural activity yield much better resolution (Yu et al 2012). While nanofibers are clearly the superior technology, there are still complications with large-scale manufacturing at the nanoscale.
When decoding the images of neural activity we correlate them to some end goal such as raising one’s arm. To consistently send the proper signals to initiate the specific neural activity that controls the execution of the specific goal takes a lot of practice on the part of the patient. Unlike other parts of the body, the hands and arms require some of the most complex prosthetics due to the complicated tasks are hands often have to perform. Even simply putting on and taking off a hat is an exceptionally complex action. Schultz et al. also discusses other ways of recording neural activity. Mentioned approaches include the use of myoelectric interfaces, measuring the electrical activity of the muscles (myo-) from outside the body or electrocorticograms/electroencephalograms, which measure activity either over the cortex or over the scalp, respectively (Schultz et al. 2011). The downside of these other approaches is that while they are less invasive in some cases, the lower resolutions or lack of specificity result in devices that require more training as the patients adjust to activating different muscles to control their prostheses. With all of these applications we can see how technology of the current day and age has revolutionized the ancient idea of prosthetics.
Beyond the different ways of recording neural activity, neural interfaces are also used to promote neural activity through electrical stimulation. In these ways, the goal of using neural interfaces and prosthetics isn’t just to replace function, but can also be to rehabilitate the patient and improve the patient’s natural function. He et al focus on the use of neurointerfaces for prosthetics of the lower body, specifically for those who have suffered from Spinal Cord Injury (SCI). The spinal cord is the major highway along which all of the motor and sensory neurons run to communicate between the body and the Central Nervous System. In a lot of prosthetics, the neural interface works as a bypass by which the signals of the nervous system can skip over the injured section of the spine. These prosthetics also use electrical stimulation on the injured section of the spine. In general, the nervous system is extremely plastic. What this means is that it is very capable of changing, creating new connections and pathways or altering existing ones. With this idea the electrical stimulation is used in hopes that the plasticity of the supraspinal, below the spine, structures can re-form to regain function lost after SCI (He et al 2008). This cause and effect of stimulation and growth isn’t specific to the supraspinal region. Stimulation can also occur elsewhere in the Peripheral Nervous System by sending somatosensory feedback. This stimulation, simimlarly, helps strengthen pathways leading to the site being stimulated as signals are flowing back and forth Other types of stimulation include transcranial Direct Current Stimulation (tDCS) and repetitive Transcranial Magnetic Stimulation (rTMS). It has been shown that with these two methods, changes to cortical neural pathways can be induced (Wang et al. 2010).
The field of neurointerfaces still faces many challenges especially in making the control of prosthetics or rehabilitation easier. Many recent advances are being made in the way we translate the signals of the body’s nervous system, whether it’s by measuring electrical activity at individual cells are measuring overall activity in the cortex (Aziz et al. 2005, Yu et al. 2012, Schultz et al. 2011). We’ve also made advances on the application of neural interfaces for the purpose of both replacement (prosthetics) and for the purpose of rehabilitation. The research in the direction of prosthetics is very diverse to match the diverse connections and pathways found in the human nervous system (He et al. 2008, Schultz et al. 2011). Be that as it may, there is still the central idea in prosthetics of using the plasticity of the nervous system and restructuring/strengthening neural pathways (He et al. 2008, Wang et al. 2010). Though our knowledge of the nervous system isn’t quite comprehensive and extensive patient training with their prosthetics is necessary, there have been many advancements in prosthetics because of our technology and growing understanding of neurointerfacing.
Aziz, J., Bardakajian, B. L., & Derchansky, M. (2005, August 7). 256-Channel Integrated Neural Interface and Spatio-Temporal Signal Processor. roc. of the IEEE Midwest Symposium on Circuits and Systems, 5075-5078. Retrieved January 25, 2014, from IEEEXplore.
He, J., Ma, C., & Herman, R. (2008, July). Engineering Neural Interfaces for Rehabilitation of Lower Limb Function in Spinal Cord Injured. Proceedings of the IEEE, 96(7), 166-1152. Retrieved January 25, 2014, from IEEEXplore.
Schultz, A. E., & Kuiken, T. A. (2011, January 1). Neural Interfaces for Control of Upper Limb Prostheses: The State of the Art and Future Possibilities. American Academy of Physical Medicine and Rehabilitation, 3(1), 55-67. Retrieved January 25, 2014, from ClinicalKey.
Wang, W., Collinger, J. L., Perez, M. A., Tyler-Kabara, E. C., Cohen, L. G., Birbaumer, N., & Brose, S. W. (2010, February 1). Quality of Life Neural Interface Technology for Rehabilitation: Exploiting and Promoting Neuroplasticity. Physical Medicine and Rehabilitation Clinics of North America, 21(1), 157-178. Retrieved January 25, 2014, from ClinicalKey.
Yu, Z., McKnight, T. E., Ericson, M. N., Melechko, A. V., Simpson, M. L., & Morrison, B. (2012, May 1). Vertically aligned carbon nanofiber as nano-neuron interface for monitoring neural function. Nanomedicine: Nanotechnology, Biology, and Medicine, 8(4), 419-423. Retrieved January 25, 2014, from ClinicalKey.