Putting the Chemical in Electrochemical Brain Interfaces
|
This is a Printer Friendly Article on the Virtual
Worldlets Network. Return to the web-view version. |
Additional Information
|
|
This is a Printer Friendly Article on the Virtual
Worldlets Network. Return to the web-view version. |
Additional Information
|
Neroprosthetics, prosthetic devices either on the surface of, or implanted deep into the brain, are a wonderful interface type. Capable of bridging damaged sections of the brain, or of calming unwanted runaway electrical patterns when used for medical reasons, or capable of direct thought-based control of computer systems when used as an interface.
You think it, and the computer does it, in essence. The neuroprosthetic doesn't even have to understand what you are thinking about; it senses the electrical pattern in your brain when you think of a given concept, or the spike known as an event related potential when your senses encounter new data. A computer system interprets this data and looks for a pattern match to an expected command.
Ultimately, neuroprosthetics, of both invasive and non-invasive types are expected to be the ultimate interface for a virtual reality system. Bypassing the body almost completely, sensory information can be replicated and sent into the brain via the brainstem or the cranial nerves by prosthetics designed for the task of providing falsified information for the senses. The brain's control signals to the body can likewise be intercepted, and in theory the originals suppressed so the body does not act, but a VR based avatar does.
This type of interface – brain machine direct interfaces, are being developed faster than ever before, with more and more research funds poured into them. It almost seems like the golden age will be here any year.
Almost.
Aside from the physical problems inherent in placing a prosthetic deep into the brain, and the damage to surrounding tissue that results, there is one rather large, and rather serious problem with the current approach all brain machine interfaces use.
All interfaces are electrical in nature. Picking up the electrical impulses fired between the dendrites of neurons in the brain and in the peripheral nervous system en route to and from the brain. This works fine. It would work perfectly if the brain's computation worked solely on electrical engrams.
Unfortunately, the brain is not electrically based. It is electrochemical. Chemical secretions between cells in the brain carry a great deal of information. They often regulate how the electrical information will be transmitted. So, by only monitoring the electrical side of communication, our interfaces are missing half the communication that goes on, right off the bat.
|
The only drawback to current BMI systems is that they are based on electrode arrays detecting or instigating electrical activity. This works fine for tasks such as suppression of epileptic seizures, or for interfacing with the peripheral nervous system, where transport of electrical neural codes is the order of the day. However, the brain itself is electrochemical, and if we don't monitor the chemical neurotransmitter signals, we are missing half the conversation.. |
It is a rather unconventional problem, and requires an unconventional solution. One such potential solution may have been found by a research team who ironically were not studying the brain at all; they were studying bacteria colonies.
A research team from Columbia University, has been trying to find a way to study signalling in bacteria colonies. Such colonies often act as a unit, each creature part of a larger structure than itself, with signs of strong organisation and communication across the colony. However, bacteria aren't like neurons. They don't press dendrites to one another, to let electrical signals jump across microscopic gaps. They are individual creatures, each cell a different being, and they communicate via chemical secretion.
It's the other half of the puzzle directly. To understand and track these secretions, requires a computer chip designed to detect and analyse these chemical secretions in situ. Research team leads Ken Shepard a biological and electrical engineering professor, and Lars Dietrich a biological systems associate professor, lead a team of PhD students to try to find a way to image the signalling molecules in real-time, and build up a comprehensive picture of the signalling behaviour within bacteria colonies.
To this end they have developed a chip based on complementary metal-oxide-semiconductor (CMOS) technology that enables them to electrochemically image the signalling molecules from these colonies spatially and temporally.
It is such an off the wall application, that to the best of anyone involved's knowledge as well as our own, this is the first time anything like this has been even attempted, let alone achieved.
The team believed there might be a way to track chemical secretions of the bacteria cells in real-time by using techniques that employ direct transduction to electrons, without using photos as an intermediary. They made an integrated circuit, a chip that, Shepard says, is an ‘active’ glass slide, a slide that not only forms a solid-support for the bacterial colony but also ‘listens’ to the bacteria as they talk to each other.”
In their studies, the team looked specifically at phenazine compounds, which are secreted metabolites that control gene expression. This matched perfectly with their intent of course; to track how the bacteria of any given colony or species modify one another's behaviour chemically. However, the technique should be applicable to any other molecules present in a biological system.
A mammalian brain is of course a much, much more complex environment than a bacteria colony, and the number of protein molecules in such an environment is many, many orders of magnitude greater. Close to a hundred common neurotransmitter chemicals have been identified in the human brain, which directly alter the way the electrical signals are transmitted.
The researchers are continuing their research by attempting to develop a chip much more intricate than the prototypes they have. They essentially have a proof of concept that detects and tracks the distribution of one class of molecules across a largely two dimensional cellular environment. The next generation will be larger and possessing a tighter internal circuitry to allow increasingly accurate resolutions of larger and larger areas.
It is a long, long way from a massively multi-chemical trail detector, but there is no denying it is on the same road. For the first time, it is a realistic possibility that one day we will be tracking and interfacing with the chemical signals in the brain, as we do with the electrical signals.
Then we will truly have a complete mind-machine interface. Until such times, we have the knowledge that it is at least possible to track the chemical half of an electrochemical system in situ, at an individual cellular resolution, and without disturbing the surrounding cells.
How we would go about integrating such a detector system into the 3D environment of the neural towers in parts of the human brain, is something that can wait. At least until these new CMOS systems are integrated with mainstream lab on a chip technology to increase the number of chemicals they can detect.
CMOS Technology Provides New Insights into How Biofilms Form
Integrated circuit-based electrochemical sensor for spatially resolved detection of redox-active metabolites in biofilms (Paper, Paywalled)
