Our goal is to leverage Google expertise and resources to advance understanding of the structure and function of the brain.


Neurons are cells in the brain that connect and communicate with each other. This neuron-to-neuron connectivity defines a network, and a major hypothesis in modern neuroscience is that structure of these brain networks can be linked to their function -- how the brain encodes memories, extracts features from perceptual stimuli, and makes decisions.

The structure of these brain networks has remained largely unknown, due to technical difficulties involved in imaging and reconstructing the brain in 3D. A diversity of new microscopy techniques has begun to address the challenge of imaging the brain in 3D at nanometer resolution, but these new machines have led to a huge bottleneck in the subsequent step of data analysis. Our goal is to help solve some of these data analysis problems and thus enable a high-throughput approach to studying the network architecture of the brain.

Initial work has focused on reconstruction of neuronal wiring diagrams from 3D electron microscopy images of neural tissue. In this approach, the brain is imaged at high resolution (e.g., 8x8x8 nanometers) with an electron microscope, and then several image analysis problems need to be solved in order to generate the wiring diagram of the portion of the brain that was imaged: aligning consecutive 2D image slices into a coherent 3D volume, tracing wires to their parent cell body, and detecting and characterizing synapses. Due to the vast amounts of data involved -- a single cubic millimeter of tissue may generate a thousand trillion pixels of image data -- computational methods that automate these tasks are an urgent necessity.


Reconstruction of a portion of zebra finch brain. Colors denote distinct objects in the segmentation that was generated using a flood-filling network, a new approach to automated neuron reconstruction developed at Google AI, in collaboration with the Max Planck Institute for Neurobiology. Gold spheres represent synaptic locations automatically identified using a previously published approach.

Featured publications

Open-source software

We believe in providing open-source software to aid in scientific research, and to further the study of the network architecture of the brain.

People involved


We are collaborating with the the Max Planck Institute, HHMI Janelia Research Campus, Harvard University, and other organizations in order to develop technology for mapping brain circuitry at synaptic resolution.

Join us

We're looking for talented people to join our team