Forgot your password?
typodupeerror
Medicine Technology

New Imaging Method Reveals Brain Connections 95

Posted by samzenpus
from the look-at-the-big-brain-on-you dept.
An anonymous reader writes "Researchers at the Stanford University School of Medicine, applying a state-of-the-art imaging system to brain-tissue samples from mice, have been able to quickly and accurately locate and count the myriad connections between nerve cells in unprecedented detail, as well as to capture and catalog those connections' surprising variety. A typical healthy human brain contains about 200 billion nerve cells, or neurons, linked to one another via hundreds of trillions of tiny contacts called synapses. It is at these synapses that an electrical impulse traveling along one neuron is relayed to another, either enhancing or inhibiting the likelihood that the second nerve will fire an impulse of its own. One neuron may make as many as tens of thousands of synaptic contacts with other neurons, said Stephen Smith, PhD, professor of molecular and cellular physiology and senior author of a paper describing the study, to be published Nov. 18 in Neuron."
This discussion has been archived. No new comments can be posted.

New Imaging Method Reveals Brain Connections

Comments Filter:
  • by AndyFewt (694753) on Wednesday November 17, 2010 @07:41PM (#34262756)
    So they have this wonderful new imaging method that can show something unseen until now... and they have no pictures with the article.

    Seriously?!
  • First (Score:1, Offtopic)

    by Stooshie (993666)
    Researchers at Stanford imaged my brain so quickly that they got me to make this first post remotely.
  • ...on my cryonically preserved brain?
    • The Stanford team used brain samples from a mouse that had been bioengineered so that particularly large neurons that abound in the cerebral cortex express a fluorescent protein, normally found in jellyfish, that glows yellowish-green."

      Not unless you have been genetically modified to have jellyfish proteins in your brain.

    • Re: (Score:2, Funny)

      by sharkey (16670)
      "Now that brain that you gave me. Was it Hans Delbruck's?"


      ........'No".
      • Re: (Score:2, Funny)

        by eriqk (1902450)
        Ah. Good. Uh... would you mind telling me... whose brain... I did put in?
  • by MichaelSmith (789609) on Wednesday November 17, 2010 @07:50PM (#34262848) Homepage Journal

    A slab of tissue — in this case, from a mouse's cerebral cortex — was carefully sliced into sections only 70 nanometers thick. (That's the distance spanned by 700 hydrogen atoms theoretically lined up side by side.) These ultrathin sections were stained with antibodies designed to match 17 different synapse-associated proteins, and they were further modified by conjugation to molecules that respond to light by glowing in different colors.

    In case you were wondering, you have to be dead to be scanned with this technique, and it doesn't look like they will be able to press a button and scan a whole brain.

    • by rubycodez (864176)

      no problem, I have a list of politicians we can scan. Also, my boss should be scanned.

      • no problem, I have a list of politicians we can scan. Also, my boss should be scanned.

        It wont work. The politicians don't have a brain.

        • Re: (Score:3, Funny)

          by wierd_w (1375923)

          No, they have brains, they just exhibit below average activity in the cortex, and above average activity in the limbic system.

          That's why they think that any measure that is intended to "Protect the children" is OK.

          (Consequently, it is also why they spend such an inordinate amount of time and resources chasing people around bathroom stalls and cloak rooms.)

      • by gmuslera (3436)
        Not sure about your boss, but won't be good to scan those politicians. Would be inhuman to force machines to do that.
    • It also requires a specimen that is genetically engineered to have fluorescent neurons. So doing this on humans at all is out of the question.
      • Makes me wonder if you could retrofit synapses with something which emits photons when action potentials change. I have molecules of Carbamazepine [wikipedia.org] in most of my sodium channels for example. It gums up the works somewhat (thats the point of taking the drug) but if you could find something which diverts a bit of energy to make photons then you may be able to extract information in real time.

        • That's not a bad idea. The biggest problem action potentials are in the range of ~40 mV, which means that the photons generated would have to be very low energy and wouldn't be able to be detected through the tissue. But you might be able to do something like a PET scan where the tracer is selectively attracted to active neurons and the emission is from radioactive decay.
    • Re: (Score:2, Funny)

      by G3CK0 (708703)

      A slab of tissue — in this case, from a mouse's cerebral cortex — was carefully sliced into sections only 70 nanometers thick. (That's the distance spanned by 700 hydrogen atoms theoretically lined up side by side.) These ultrathin sections were stained with antibodies designed to match 17 different synapse-associated proteins, and they were further modified by conjugation to molecules that respond to light by glowing in different colors.

      In case you were wondering, you have to be dead to be scanned with this technique, and it doesn't look like they will be able to press a button and scan a whole brain.

      I'm not so sure that this small technicality will stop the TSA from installing one of these scanners :)

    • by rdnetto (955205)

      If they can get this working, you might not remain dead. As I understand it, all the information in a neural network is stored in the connections between neurons - that is, how much of the signal gets transmitted to the next neuron. This is usually represented as a decimal between 0 and 1 in programming.
      If we could measure those values and record the arrangement of neurons, it would be possible to literally copy someone's mind. Actually running it would require a ridiculously powerful computer, but Moore's

      • You only need a powerful computer if you want to run the simulation in real time. From the point of view of the simulation that may not be necessary. I have my doubts about reconstructing the personality from a static analysis of the brain. A lot of the information transfer is serial so knowing what is connected to what doesn't tell you enough about the protocols they are using.

        • by rdnetto (955205)

          You would need the speed to be at least within an order of magnitude of realtime for it to have any practical use. Within the brain/neural network whatever protocols are in use is irrelevant, since it's communicating with itself. As for connecting it to reality, connecting the I/O neurons to another dead/cloned body should suffice.
          Really, the only issue I can think of is that if the values of the connections couldn't be changed by the 'mind', then the person would be unable to learn anything new and would t

      • by tehcyder (746570)

        If we could measure those values and record the arrangement of neurons, it would be possible to literally copy someone's mind.

        What about Heisenberg's Uncertainty Principle? Are you really sure we will ever be able to capture tens of billions of events simultaneously with perfect accuracy?

        • by rdnetto (955205)

          This is on the cellular level, so I don't think that applies. 'Capturing tens of billions of events simultaneously with perfect accuracy' is a strawman - we don't need observe all the signals simultaneously, we just need to measure the resistance (for lack of a better word) along each connection. 'Perfect accuracy' isn't found anywhere in science - we just need it to be good enough.
          A century ago, people would never have believed it possible to communicate across the planet in under a second, or to communica

  • Hopefully this will lead to further breakthroughs in biometric prosthesis. If they can map out where the nerves are and what their functions are more accurately, we may soon be able to interface with them more directly. Imagine a prosthetic arm that actually has feelings versus our current ones that only have motion. This could a a very good thing... or scary for those afraid of cybernetics.
    • Imagine, a therapy by which one can backup their brain. Lost Cell Phones, Keys, Girl Friends birthdays. The last one alone is worth the research.
      • by tehcyder (746570)

        Imagine, a therapy by which one can backup their brain. Lost Cell Phones, Keys, Girl Friends birthdays. The last one alone is worth the research.

        If you've forgotten your girlfriend's birthday, having a backup of your brain would be shutting the stable door after the horse has bolted, although it would be useful if it could be transplanted into a new body after she'd tortured the old one to death.

  • Are these the human-brained mice of which I've heard so much of late?

  • by Casai (1011763) on Wednesday November 17, 2010 @08:17PM (#34263036)
    This is immunohistochemistry [wikipedia.org], just scaled up to many different antibodies for the same sample and realigned in space.

    Also, the connectivity is lost. You can't tell which neurons are connected to which other neurons. The overall circuitry, essential for the functioning of neural networks, is invisible. All you can see is points of contact between neurons.

    Perhaps combining this technique with super high resolution diffusion tensor imaging would be a way forward. Although, as far as I know, DTI is nowhere near neuron or axon resolution as of yet.
    • Re: (Score:2, Interesting)

      by Casai (1011763)
      Also, I'd be interested to see how (or if) they managed to completely wash off antibodies between scans without damaging the tissue or disrupting synaptic structure. Many synaptic proteins recognize and bind each other in the same way that antibodies bind their antigens, so it stands to reason that disrupting antibody binding would also disrupt the binding of these proteins.
    • by DeadCatX2 (950953) on Wednesday November 17, 2010 @08:41PM (#34263274) Journal

      I was just about to come here and mention DTI, but you beat me to it.

      I'm not sure if they're down to neuron/axon resolution yet, but I do know they're pretty close. Dr. Walter Schneider at the University of Pittsburgh has created a movie image of the various connections in his brain.

      http://www.lrdc.pitt.edu/schneider/ [pitt.edu]

      • Re: (Score:3, Informative)

        by dogmatixpsych (786818)
        I do this exact research (diffusion weighted imaging of human brains). We are no where near neuronal/axonal resolution with diffusion weighted scanning (DTI is a special case of diffusion weighted scanning - there are better methods than DTI for analyzing images: e.g., http://brainybehavior.com/neuroimaging/2010/08/hardi_vs_dti/ [brainybehavior.com]).

        With live humans we only resolve down to about 2mm^3. There are many neurons and axons in that space. At best for the whole brain we create only a few fibers for that 2x2x2 mm ar
        • by DeadCatX2 (950953)

          Thanks for the info! I guess I should have better qualified "close", but it's not really my field, I just know a guy who knows a guy who... After reading your link, I think I vaguely remember something about the DTI being unable to resolve fibers that cross.

          I'm guessing that you do your studies on a 3T...do you know how much better a 7T might be?

          What resolution must we reach in order to resolve a individual fibers?

          • Yes, we use a 3T. Theoretically 7Ts would be better (there aren't any 7T scanners for people that I know of) but there are some issues that engineers are still working through (signal, noise, safety, etc.).

            To reliably resolve individual axons, we'd have to have a resolution of under 5 m (as I said earlier, we typically resolve 2000 m in vivo). That's a huge volume difference (125 cubic m vs. 8,000,000,000 cubic m)! We "take pictures" of the brain using voxels (volumetric pixels), so the 3D resolution is i
            • by DeadCatX2 (950953)

              Safety, yes. I bet the heating issue would be much worse at 7T. A google turns up a few places that have a 7T

              Okay, so we are waaay farther from fibers than I thought. Still, though, fascinating little discussion, thanks! I love talking to brain researchers for some strange reason.

              • You're welcome. You are correct, there are a few 7T human scanners being used for research now. I hadn't checked on 7T scanners for a couple years so I guess things have changed! I'd love to get my hands on a 7T scanner. :)
              • I should add that 7T MR scanners are not available commercially. This means you have to have a research agreement with the manufacturer to have one (you essentially test it out for them).
      • by ceoyoyo (59147)

        There are very convincing theoretical reasons why DTI can never resolve individual axons, at least not in a non-cryogenic sample. If you want to look at individual axons you're far better off doing it the old fashioned (and considerably less sexy) way: with a microscope.

    • There are existing techniques that give ~tens of nanometers resolution using fluorescence microscopy (discussed in a feature in Nature Methods [nature.com] ). Techniques such as PALM/FPALM/STORM (developed by Betzig, Hess, and Zhuang, independently) use photoswitchable fluorophores to image and localize single fluorescent molecules with high precision then reconstruct the image from these single molecule images. Another technique, STED (stimulated emission depletion, developed by Hell) uses stimulated emission to eff

  • "Researchers ... have been able to quickly and accurately locate and count the myriad connections between nerve cells in unprecedented detail, ..."

    Zuckerberg is working on an API for this right now.

  • by B1oodAnge1 (1485419) on Wednesday November 17, 2010 @08:30PM (#34263158)

    From the CNET article: [cnet.com]

    They found that the brain's complexity is beyond anything they'd imagined, almost to the point of being beyond belief, says Stephen Smith, a professor of molecular and cellular physiology and senior author of the paper describing the study: "One synapse, by itself, is more like a microprocessor--with both memory-storage and information-processing elements--than a mere on/off switch. In fact, one synapse may contain on the order of 1,000 molecular-scale switches. A single human brain has more switches than all the computers and routers and Internet connections on Earth.

    This is why I am extremely skeptical of claims that we will be able to effectively model the brain, or recreate it artificially, any time soon.

    • Re: (Score:3, Insightful)

      by urusan (1755332)

      The good news is that it shows how much we can still improve microprocessor technology. Perhaps Moore's Law (or something similar) will keep up for quite some time into the future?

    • by mcrbids (148650)

      This is why I am extremely skeptical of claims that we will be able to effectively model the brain, or recreate it artificially, any time soon.

      Well, maybe, maybe not. It may be that we are finding unexpected complexity in the brain. On the other hand, we're seeing continued advances in computing technology - I remember reading about the end of Moore's law 20 years ago! Sure, physics have forced certain ideas to be tossed, such as the Uni-Processor model, but progress is still occurring at a blistering pace, with my battery-powered Android phone handily outperforming the top-of-the-line computers highlighted in those "Moore's law is almost dead" ar

      • You're point is a good one, and it's why I wouldn't say it's impossible.

        However if my understanding of the article is correct we are looking at much higher than 100x the complexity that has been thought up till now, and that's just the hardware, we're even further from understanding the software.

        Again, I agree that technology is amazing and rapidly moving, but a working model of a human brain is something that I hope to live to see, rather than something I expect anytime soon. I'm 23, if they can accomplish

        • by tehcyder (746570)

          Again, I agree that technology is amazing and rapidly moving, but a working model of a human brain is something that I hope to live to see, rather than something I expect anytime soon. I'm 23, if they can accomplish that in 70 years I will be impressed and excited on my deathbed. :-)

          The point is, if they could indeed faithfully copy the human brain, you wouldn't have a deathbed. That is why it seems so unlikely to all but Ray Kurzweil fans.

    • by sempir (1916194)

      Wait till they get all this brain shit sorted out and claim they can effectively model the brain....THEN i'm gonna say "right.....now do the same with a womans brain"...Ha!....fucked....arn't you!!!

  • I need an upgrade desperately.

    • I'm not looking forward to when we can synthetically reproduce and upgrade our brains with new computing-like technologies. I don't want to have to pay millions of dollars for getting songs stuck in my head.

  • Is this going to replace the back scatter scans as the new TSA scan of choice?
  • but.. pics or it didn't happen. Thx.
  • I wonder if they imagined we would be impressed by this "finding".
  •           Pics or it didn't happen

  • http://en.wikipedia.org/wiki/Connectome [wikipedia.org]
    which is a map of the neural connections in the brain.

    I highly recommend watching this vid, demonstrating the "New Imaging" methods, its also quite humorous.
    http://www.ted.com/talks/lang/eng/sebastian_seung.html [ted.com]

  • The summary only mentions electrical pulses, it should have mentioned that the local chemical environment is part of the information exchange.

Theory is gray, but the golden tree of life is green. -- Goethe

Working...