Google Research says AI techniques are helping accelerate brain mapping by speeding up the identification, classification, and visualization of neurons. In a new paper, “MoGen: Detailed neuronal morphology generation via point cloud flow matching”, to be presented at ICLR 2026, the team uses synthetic neural shapes to improve AI reconstruction models.
The work builds on recent connectomics projects, including a complete map of the male fruit fly brain and central nervous system, which the source describes as a foundational resource for studying how the brain responds to stimuli and controls the body. That map includes 166,000 neurons and was the result of years of work by AI-enabled computers and human experts.
Google Research says reconstructing the full brains of mammals, and humans, remains far out of reach. A complete mouse brain is a thousand times larger than the fruit fly brain map, and a human brain is a thousand times larger than that.
The company says it and its partners have also mapped fragments of a zebra finch brain, a whole larval zebrafish brain, and a small fragment of the human brain, and recently launched an effort to map a small section of mouse brain.
According to the source, adding synthetic examples from the Neuronal Morphology Generation model, or MoGen, leads to a 4.4% reduction in reconstruction errors. The company says that at the scale of a complete mouse brain, this would translate to 157 person-years of manual proofreading saved.
The Google Research Connectomics team says the work adds to its growing list of foundational tools for modern neuroscience, developed over more than a decade of collaborative brain science research.
Source: research.google.
Companies can share verified announcements through Newz9’s international press release submission page.
