Scientists from Harvard University, in collaboration with Google, have developed a detailed diagram of a small fraction of the human brain, bringing new complexities to light.
The study focused on a fragment of brain tissue from a 45-year-old woman diagnosed with epilepsy, and the results were published in the journal Science.
Using advanced Google technology, the team mapped the complex neural network within a cubic millimeter of healthy cortex tissue, finding 57,000 cells, 150 million neural connections, and 23 cm of blood vessels.
This brain map has been made freely available for use by the scientific community.
Jeff Lichtman, leader of the study and professor at Harvard, explained that the challenge of this study lay in obtaining high-resolution images of the brain, an extremely complex task until then.
To do this, they used techniques for slicing tissue into ultra-thin layers and electron microscopy, capturing images on a nanometer scale.
A machine learning algorithm was used to follow the path of the cells, a process that would have taken years if it had been done manually.
Detailed analysis revealed intriguing features, such as pyramidal neurons with symmetrically branched dendrites, and axons forming loops before heading to their connections.
According to The Harvard Gazette, the images generated amounted to 1.4 petabytes of data, equivalent to 14,000 movies in 4k resolution.
Lichtman pointed out that many of the findings challenge existing knowledge and are not yet fully understood, suggesting a vast field still to be explored between current knowledge and what remains to be discovered.
For example, scientists have seen that some neuronal connections are so strong that they can explain how certain learned behaviors are executed automatically.
While more than 96% of the axons formed a single connection with a target cell, some established multiple connections, reaching more than 50 with a single cell.
Scientists have no plans to map a complete human brain due to technological challenges and a lack of healthy raw material. Current findings already challenge previous knowledge and expose unknown facets of the brain.
The next step includes a collaborative effort between several universities and Google to map the brain of an entire mouse.