Scientists Develop Brain-Microchip Bridge 118
dreampod writes "Canadian scientists have developed a microchip capable of monitoring the electrical and chemical communication channels between individual neurons. This is the first time scientists have been able to monitor the interaction between brain cells on such a precise and subtle level. In addition to providing the ability to see more easily the impact of drugs on various mental disorders during testing, this provides one of the first fundamental steps towards real mind-machine interface."
But not in a real brain? (Score:4, Informative)
TFA is vague but it looks like the cells in question are being kept alive outside the organism. I suppose this could be adapted into an implantable device, but cochlear implants almost do that anyway.
Read the small print (Score:5, Informative)
Before jumping on this, read the small print.
They take out a piece of brain tissue, and implant it into the machinery, not the other way around. I'm not sure about you guys, but that kind of interface doesn't seem too useful to me, although it could be useful for diagnosis.
Re:But not in a real brain? (Score:3, Informative)
Re:But not in a real brain? (Score:4, Informative)
Cochlear implants go the opposite direction. Cochlear implants are like speakers, this is like a microphone.
Thats true but the important thing here is the interface, which works both ways. This device may have more resolution though, and it seems precise enough to talk to individual neurons, rather than nerve cells.
Re:Read the small print (Score:3, Informative)
If there were medical engineers they could take a device like this and package it for implantation.
There is such a field as Biomedical Engineering [wikipedia.org].
Re:But not in a real brain? (Score:5, Informative)
Re:But not in a real brain? (Score:3, Informative)
Sorry, they're the same thing. Neuron == nerve == nerve cell. A neuron consists of a cell body (the prokaryon), one axon (outgoing signal), and one or more dendrites (incoming). They connect to each other from axon to dendrite, at links called synapses. The signal is propagated by very high-resolution, high-frequency balancing and shifting ion gradients.
Re:But not in a real brain? (Score:3, Informative)
Re:Chemical dialogue (Score:1, Informative)
The device in question does not measure chemical signals directly. It measures the flow of ions "in" and "out" (through the membrane) of individual neurons. When one neuron communicates with a second neuron, typically the source neuron releases a chemical (neurotransmitter) onto the target neuron, where receptors sense the chemical and in response open ion channels (pores in the membrane). Ions (such as sodium, potassium, chloride, calcium) then pass through these channels, changing the potential difference across the target neuron's membrane. This fluctuation in membrane potential then propagates to other parts of the target neuron (passively, as well as actively with the help of other ion channels that sense the change in membrane potential).
Many such fluctuations in membrane potential happen frequently in each neuron, and when they occur in the right combinations, they add up sufficiently to cross a threshold value (which varies from cell to cell) - once this threshold potential as been reached across the membrane in the main part of the neuron, a runaway cascade of further channel opening and closings is set in motion, resulting in a sudden, large spike in membrane potential. Such spikes are called "action potentials". They are essentially "digital" - they have a stereotypical time course, or shape, and are thought to encode information primarily by their existence (and not by their exact height or width). Like digital music (MP3), the advantage of such spikes is that no information is lost as they propagate to other parts of the cell, even though the spike is somewhat attenuated - as long as it is still recognized as a spike, it is still a "1", not a "0".
When such a spike reaches the end of an axon (there is one axon, but it may have many branches), the arrival of this large and quick membrane potential fluctuation causes chemicals (neurotransmitters) to be released, which then may be sensed by receptors on a different neuron's membrane. And so we have reached the beginning (or end) of another cycle, and this is how the brain transmits and processes (most) information.
The connection between two neurons is called a synapse, and is formed between the end of the axon from the source neuron and the (usually dendritic) membrane of the target neuron, which contains receptors. The source neuron is termed "pre-synaptic" and the target neuron "post-synaptic".
Currently the most accurate way to measure the electrical state of a single brain cell is a technique called "whole-cell patch-clamp recording". This involves making contact between the tiny (about 2 micrometer diameter) tip of a glass pipette and the membrane of a neuron, forming a tight seal between these surfaces (usually with the help of suction, actually drawing part of the flexible membrane into the tip of the pipette), and finally rupturing the bit of the membrane within the tip. Now the inside of the pipette is contiguous with the inside of the neuron. Ionic solution in the pipette disperses into the neuron, and a fine wire extending from within the pipette out to an amplifier is used to measure the flow of ions between the solution inside the neuron+pipette and an electrode in the solution surrounding the brain cell (the "bath"). Achieving such a measurement configuration is delicate and difficult - just how difficult depends on the sample being studied: dissociated neurons (a soup), cultured neurons (a soup that then was allowed to regrow connections), neurons in thin slices of brain tissue, or intact brains of anesthetized or behaving animals.
Various devices exist that can largely automate such "patch-clamp" measurements, but only for dissociated cells. This is mainly of use when, for example, testing the effect of drugs on neurons (or other types of electrically active cells). From what I can see of the chip mentioned here, it works for cultured neurons, which is an improvement. Perhaps something similar will, in the future, work for intact brain tissue - this is probably still years away. I find it more likely that such