A collaboration between Google researchers and Harvard neuroscientists has yielded an unprecedented view into the human brain, utilizing artificial intelligence (AI) to make sense of vast amounts of data. The team combined brain imaging with AI-based image processing and analysis to reconstruct nearly every cell and its connections within a small volume of human brain tissue about half the size of a grain of rice.
Although the 3D mapping only covers a small region of the brain, it requires a monumental 1.4 petabytes (1.4 million gigabytes) to encode. This dataset, which is freely available to the scientific community, is said to be the largest ever made of human brain structure at this resolution.
The project began with Harvard researchers collecting thousands of extremely thin cross-sectional images from a donated brain sample. This small piece of healthy brain was removed during surgery on a woman with epilepsy. The otherwise discarded tissue was preserved as part of an IRB-approved study for later analysis.
Google's research team developed advanced AI tools to construct an interactive 3D model of the brain tissue. Describing just this small sample — one-millionth of the total human brain and about 3 mm long — required more than a million gigabytes of data: 1.4 Petabytes.
The sample came from a part of the cortex (gray matter) called the anterior temporal lobe. The cortex has six layers, visible in this zoomed-out view by coloring neurons according to their size and type.
The one cubic millimeter tissue sample contained about 50,000 cells and approximately 150 million synapses — points where signals cross from one neuron to another. Some neuron pairs were found to be connected extremely strongly through as many as 50 synapses, an occurrence that researchers do not yet understand.
Intriguingly, clusters of cells tended to occur in mirror-image orientation to one another were found in the reconstruction. Neurons in the brain were found to be intensely connected, with a single neuron having more than 5,000 axons arriving from other neurons to bring signals.
A surprising finding was the occurrence of “axon whorls,” looped piles of axon that were rare in the sample. In some cases, they sat on the surface of another cell. Their function remains unknown.
The research team believes there is still much to observe and understand from this reconstruction. They hope other researchers will use the data to make additional discoveries. Scientists believe that by continuing research into the brain’s connections, they can one day understand things like how our memories form or what leads to neurological disorders and diseases like autism and Alzheimer’s.