3D model of a cubic millimeter of human brain in nanometric resolution

Etxebeste Aduriz, Egoitz

Elhuyar Zientzia

Ed. Google Research & Lichtman Lab (Harvard University). Renderings by D. Berger (Harvard University)

Researchers at Harvard and Google have developed the largest 3D model of human brain with nanometric and synaptic resolution. It's a cubic millimeter of brain, roughly the size of a grain of rice, which, although it doesn't seem much, has 57,000 cells and 150 million synapses. In total, 1,400 terabytes.

From a temporal cortex sample, the model was obtained using electron microscopy images and artificial intelligence algorithms. It shows in detail almost all the cells of this brain segment and its network of neural connections, with a visible resolution of all the synaptic links.

In this model, structures never seen so far have been found, such as axon assemblies connected to 50 synapses. The sample consisted of a patient with epilepsy and researchers are not sure that these structures are pathological or simply infrequent.

In fact, researchers believe that these kinds of maps will allow us to better investigate and know the brain and brain diseases. All the information and some of the tools they've developed are available for other researchers to exploit this big data set.


Eusko Jaurlaritzako Industria, Merkataritza eta Turismo Saila