The Johns Hopkins Department of Physical Medicine and Rehabilitation (PM&R), in collaboration with the Johns Hopkins University Applied Physics Laboratory and the Department of Neurology and Neurosurgery, has been awarded a grant by the Defense Advanced Research Projects Agency (DARPA) to conduct a clinical trial focused on recording and stimulating the brain of a person with tetraplegia.
Ambarish Pawar, Research Fellow with the Department of Physical Medicine & Rehabilitation at Johns Hopkins, explains three findings published in the Journal of Neurosurgery, Brain Stimulation, and the IEEE Engineering in Medicine & Biology Society Virtual Conference.
Citations:
McMullen DP, Thomas TM, Fifer MS, Candrea DN, Tenore FV, Nickl RW, Pohlmeyer EA, Coogan C, Osborn LE, Schiavi A, Wojtasiewicz T, Gordon CR, Cohen AB, Ramsey NF, Schellekens W, Bensmaia SJ, Cantarero GL, Celnik PA, Wester BA, Anderson WS, Crone NE. Novel intraoperative online functional mapping of somatosensory finger representations for targeted stimulating electrode placement: technical note. J Neurosurg. 2021 Mar 26:1-8. doi: 10.3171/2020.9.JNS202675. Epub ahead of print. PMID: 33770760.
Osborn LE, Christie BP, McMullen DP, Nickl RW, Thompson MC, Pawar AS, Thomas TM, Alejandro Anaya M, Crone NE, Wester BA, Bensmaia SJ, Celnik PA, Cantarero GL, Tenore FV, Fifer MS. Intracortical microstimulation of somatosensory cortex enables object identification through perceived sensations. Annu Int Conf IEEE Eng Med Biol Soc. 2021 Nov;2021:6259-6262. doi: 10.1109/EMBC46164.2021.9630450. PMID: 34892544.
Christie BP, Osborn LE, McMullen DP, Pawar AS, Thomas TM, Bensmaia SJ, Celnik PA, Fifer MS, Tenore FV. Perceived timing of cutaneous vibration and intracortical microstimulation of human somatosensory cortex. Brain Stimulation. 2022 May-Jun: 881-888. doi:10.1016/j.brs.2022.05.015. PMID:35644516.
globally every year. About half a million people lose use of their limbs due to spinal cord injury. Both in terms of being able to move them and being able to feel sensations in their limbs. Brain machine interfaces or B. M. I can help these people regain function of the limbs by using brain signals to directly control prosthetic or robotic limbs. That's bypassing the damaged spinal cord. Brain machine interfaces work by reading signals directly from the brain to micro electrode arrays implanted in the hand representation area of the brain. The signals can then be decoded and used to move toward the account. Now being able to control movement of robotic on to interact with everyday objects is just one part of the equation. Receiving sensory feedback when we grasp or manipulate an object is also extremely important. For example, when we pick up a glass of water to drink, we receive a lot of sensory information including the shape that share and the weight of the glass. We use this information to plan smooth movements to bring the glass to our mouths. So being able to move the army is not enough. We need to feel the sensations to complete even very simple movements. So my work is focused on incorporating this sensory feedback in brain machine interfaces so that the robotic arm feels more like a native are we can do that by direct electrical stimulation of the area of the brain that represents the hands. Thus bypassing the damaged spinal cord. This work was done in collaboration between the department of Physical Medicine and rehabilitation at johns Hopkins and the johns Hopkins Applied physics lab here, I'm going to focus on three studies. All these studies were done in a single participant who had limited function of his arm. This participant also had close to normal sensation in the hand and the finger depression. The first study describes strategies to implant stimulation electrodes in the correct area of the brain. For example, if you want to grasp or manipulate an object, we need to provide sensory feedback to the fingers. So it's important to place the electrodes precisely in the area of the brain that has hand or finger tip representation only. Then can you achieve precise stimulation? Generally finding the right areas of the hand is done through functional M. R. I done before the surgery. This study describes a novel online functional mapping strategy that is enduring the micro electrode implantation surgery and that can potentially be better localized fingertip areas in the somatosensory cortex. What was done here was during the surgery, the participant was woken up and a large dense electra cartographic grid was placed on the cortex. Note that this was a much larger grid that covered a much larger area compared to the really tiny micro electorates that would eventually be implanted due to the size the grid was able to cover a larger area of the brain regions of interest. After placing the grid, each finger of the hand was stimulated through mechanical vibrations through more displaced on the finger, we then look for electrodes on the grid that showed strong gamma activity. Whenever the fingers were mechanically stimulated, this electoral grid with the active electrodes was then overlaid on a picture of the brain surface, based on where the active electrodes were situated. On the picture of the brain surface, we were able to precisely find the locations to implant the micro electrode race and were able to get a good representation of the fingers this way online function mapping can be used for making sure that micro electoral arrays are implanted in the correct desired location of the brain so that we can provide precise sensitive feedback in the form of brain stimulation. So now that we have been planted in the correct fingertip areas of the brain. In the second study, we investigated how long it takes between brain stimulation and when the participants perceived the sensation. This timing is very important in natural mechanically evoke sensation when our fingers judge something. The sensation is almost instantaneous, which allows us to do dextrous manipulation of objects. Now imagine if there was half a second delay between when you grasp the object and when you sense the object, you'll barely be able to do any object manipulation that way. So the question here is, can we achieve similar instantaneous sensation when we electrically stimulate the brain. So here the objective was to characterize when the brain stimulation based sensation was consciously experienced. And to compare this timing to mechanically will counterpart in our participant note that we can do this experiment because a participant has almost normal in tax sensations in his fingertips. What we found was very interesting that brain stimulation evoked sensations that was slightly slower than the mechanical evoked sensations, but not by much. This result is very important because it shows that sensations evoked by brain stimulation is fast enough to be effective in brain machine interfaces. So now that we know that electrical brain stimulation can be perceived at the fast enough time scale. The next question is, can we use this to have spinal cord injury patients to grasp objects in real life? So in the third study, the goal was to develop stimulation strategies that can enable patients to identify three different virtual objects based solely on the sensory feedback that is given to brain stimulation. Here, the participant cannot see or physically touch the object, all the information that they had was coming through sensory feedback through brain stimulation. The three version objects that we use in our study were a cylinder. The second object was a positive slope object, which was a conical which basically decreased in diameter in one direction and a negative slope object, which is also chronicle, but but it increased diameter in the opposite direction. The brain stimulation buried in time and space in space, meaning for a cylinder that has the same diameter throughout all, fingers received stimulation at the same time for the positive and the negative slope objects due to the nature of the object and how it was gripped. The timing of when the sensation was perceived between individual fingers deferred. That is some fingers got sensitive slightly before the others stimulation was also varied, such that when grass was initiated, stimulation intensity ramped up but stayed constant after the grass was completed and then gradually decreased as the object was let go. Based on this type of strategy, the participant was able to identify these three different objects more than 80% of the times. This is a very interesting and important result because it shows that this kind of brain stimulation that varies in space and time due to the nature of the object, is a viable option to provide sensitive feedback that can help patients interact with objects in their life taken together. These three studies show that brain stimulation given to the correct part of the hands can be used to give sensory feedback to spinal cord injury patients that can effectively enable these patients to dexterously manipulate objects in real life
