Scientists from Google and Harvard have mapped the human brain using 225 million individual images.

The more knowledge a person acquires, the more aware he becomes of how little he knows about the Earth, and even about the universe. There is still something to explore in the deep seas of our planet, and if all goes well, the mission to Mars could begin in the 2030s on the neighboring red planet, which is either ridiculously close to cosmic standards or a ridiculously small portion of the world. universe. . But why do we want to go to Mars while our knowledge of the human brain is too boring, even after so much research behind us? This parallel occurs between the cosmos and the brain because a joint project between Google and Harvard University (Lichtman Laboratory) has produced a 3D map of the human brain that evokes Google Earth. If you want to crash into letcho, the awesome “Google Brain” find it here. Sure, it’s spectacular, and all, but it doesn’t hurt to be aware of what this brain is. We are sure, of course, that everyone was watching biology class!

The human brain, however, weighs less than 1.5 kilograms and has a volume of just over 1 liter. It is estimated that between 50 and 100 billion neurons (neurons) in a network, of which approx. 10 billion pyramidal cells with a vivid connection between approximately 100 billion synaptic contact points and protrusions (short-dendritic, long-axon). This is why the network drawn by Google is so intricate, with almost 4,000 axons connected to a neuron.

The creators of the brain map put it together from 225 million separate images, which is 1.4 petabytes (peta is a thousand times more than tera). The aim of the project is to be able to study the human brain in even more detail, on an even larger scale, which for this project specifically consists of 86 billion neurons and 100 billion synaptic connections (so a sample of said brain was taken ).

Back in January 2020, Google launched a database that mapped half of a music’s brain, so this current project is a huge leap. It all started with a microscopic sample taken from a cortical temporal lobe: this was stained to make it removable and then coated with resin to protect and preserve it. It was then cut into 5,300 cubes, each about 30 nanometers thick. An electromicroscope with scanning capacity capable of recording up to a resolution of 4 nanometers was used to obtain images. This was used to create the 225 million 2D images that were combined to achieve the 3D effect. Of course, intelligent machine learning algorithms were also used in the process, thanks to which it was possible to separate cells and structures into different slices.

Google has christened this map the HO1 dataset, and we can see 50,000 cells and 130 million synapses in it. You can also see the project in a video below and you can read a more extensive study about it here, from which the image used for the article is also derived.

Article Source: Syfy.