A collaboration between the research team of Google and Harvard University made the three-dimensional map most detailed ever of every cell and every cellular connection in a human brain sample approximately 3 mm and with a volume of 1 mm3containing approx 57,000 cells And 150 million connections. This is an unprecedented view of our brain obtained through a combination of techniques brain imaging processed by artificial intelligence algorithms developed by the Mountain View giant.
The sample was taken during surgery on a woman suffering from epilepsy: During the operation, doctors had to remove a small portion of healthy cerebral cortex, which was preserved and became research material for Google.
The map has a “weight” of a lot 1.4 petabytes (1.4 million gigabytes) and reproduces details on the order of thousandth of a millimetre. This makes it the largest and most informative dataset ever created of a portion of the human brain at this resolution. Google has made the map available to the entire scientific community, who will be able to use it for neurological research: Possible applications can range from studying memory to understanding the neurological causes of disorders such as autism or Alzheimer's disease.
What Google 3D map images show
The brain sample was taken from the cortex of the anterior temporal lobe. To map it, the researchers produced thousands of images of extremely thin sections of the sample, which were then analyzed and linked using Google's AI algorithms. In this image of a detail of the sample, for example, the different layers of which the cortex is composed can be distinguished through the colors that indicate neurons of different types and sizes.
Going into more detail, you can see the individual brain cells and the synapses they create in impressive detail. In this image for example we can observe a group of excitatory neurons with dimensions varying between 15 and 30 thousandths of a millimetre. The colors reflect their size: red for the largest, blue for the smallest.
Going to the scale of the single neuron, the researchers also identified groupings of cells with mirror configurations of each other, as in the particularly symmetrical example below.
One of the most interesting aspects of the map is the possibility of analyzing the complex system of connections between brain cells. In this image, for example, we are looking at a single neuron with the map of all about i 5000 axons (in blue) that connect it to other neurons and cells synapses (green) that carry signals from axons to other neurons.
Going into even more detail, the map allowed the researchers to discover that in some rare cases, the axons (in blue in the image below) form tangled structures sometimes supported by another neuron (in yellow). It is currently unknown what the function of this particular configuration is.
Axons can activate a neuron (that is, tell it to send a signal) or deactivate it. In this image we see a neuron (in white) surrounded by the axons capable of activating it (in green) and those capable of “turning it off” (in blue).
