backtop


Print 5 comment(s) - last by Zo0noUno.. on Mar 14 at 2:04 AM


  (Source: topnews.in)
Could be useful for those with neurological conditions

Harvard researchers have created a 3-D nanoscale model of neural circuits, which allows the medical school team to "crawl" through the neural network's individual connections. 

Clay Reid, study leader and Harvard Medical School professor of neurobiology, along with graduate student Davi Bock and postdoctoral researcher Wei-Chung Allen Lee, have developed a 3-D model capable of showing researchers individual connections in the neural network, which could provide information about the brain we never knew before.

Until now, researchers have worked to find ways of understanding the brain's inner workings, especially in the cerebral cortex. The cerebral cortex processes reasoning, sensory input, and possibly even free will, making it one of the brain's most important regions. Researchers have understood a broad outline of the anatomy of the cerebral cortex for some time, and more recently, imaging technology has allowed them to look closer into the brain's methods of processing information. While this has showed what a circuit actually does, it hasn't showed us how it operates. This is because one circuit contains 10,000 to 100,000 neurons, where each makes 10,000 interconnections. This is a total of 1 billion connections within one single circuit, and this makes it difficult to get into the circuit and examine its wiring

But Reid's team has been working at this problem for several years. They've studied the cerebral cortex thoroughly, isolating individual neurons while they "fire in response to external stimuli." Now, the team has successfully created a model capable of showing individual connections at work. 

The Harvard team injected the region of a mouse brain responsible for vision with dyes that "flashed" whenever certain neurons fired. Then, using a laser-scanning microscope, they recorded these firings. Electron microscopy was then utilized to see these neurons and hundreds of others "with nanometer resolution." A new imaging system was then used to capture over 3 million high-resolution images, which were sent to the Pittsburgh Supercomputing Center at Carnegie Mellon University to be stitched into 3-D images. Ten individual neurons were chosen from these images and their connections were traced through the brain in order to develop a diagram. 

Through this model, Reid and his team found that neurons that suppress brain activity are randomly wired, suppressing neurons in local groups all at once, which is an important find for those with neurological problems. For instance, epilepsy is caused by neural restriction gone amiss.

"This is just the iceberg's tip," said Reid. "Within ten years I'm convinced we'll be imaging the activity of thousands of neurons in a living brain. In a visual circuit, we'll interpret the data to reconstruct what an animal actually sees. By that time, with the anatomical imaging, we'll also know how it's all wired together."

Reid plans to keep working to produce larger data sets, and to perfect the overall model.

This study was published in Nature.
 


Comments     Threshold


This article is over a month old, voting and posting comments is disabled

Neuro Net
By finbarqs on 3/11/2011 9:00:08 PM , Rating: 3
my CPU is a neuro network processor, a learning computer..




RE: Neuro Net
By captchaos2 on 3/13/2011 4:31:56 PM , Rating: 2
I can't believe that they didn't put a picture of a wireframe Max Headroom from the tv show on this article! The dvd collection was nice!


"It's okay. The scenarios aren't that clear. But it's good looking. [Steve Jobs] does good design, and [the iPad] is absolutely a good example of that." -- Bill Gates on the Apple iPad

Related Articles













botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki