Researchers have for the first time created a 3-dimensional (3-D) model of the human brain so precise that it could allow scientists to see microscopic details, such as single layers or sublayers of the cerebral cortex.
Currently used magnetic resonance imaging (MRI) has a spatial resolution of about 1 mm. But the new 3-D model exceeds that by about 50-fold, with a spatial resolution of 20 micrometers. “This allows us a completely new level of insight into the brain’s organization,” coauthor Alan Evans, PhD, of the Montreal Neurological Institute of McGill University in Canada, said during a teleconference describing the research. The project’s details are published online today in the journal Science.
The research team, led by Katrin Amunts, DrMed, director of the Institute of Neuroscience and Medicine at the Jülich Research Center in Germany, created the model using a deceased 65-year-old woman’s healthy brain. They sliced the brain into 7404 sections, each about as thick as a human hair. Digitized images of the histological sections generated a 1-TB dataset, which is more than 100 000 times larger than a typical MRI. The team has had to develop new software that enables them to probe such a vast amount of data.
“This… is big science comes to the brain,” Evans said. Hence, the model is dubbed “BigBrain.” The research is part of the European Brain Project, which intends to reconstruct the human brain in supercomputer models and simulations. In April, President Obama announced a $100 million US initiative to map the human brain.
BigBrain’s clinical applications include locating more precise targets for deep brain stimulation to treat Parkinson disease or Alzheimer disease, an approach currently in clinical trials. “This could serve as a new atlas for neurosurgery,” Amunts said. The model also could help researchers locate abnormalities in neurotransmitter receptors that may be linked with specific brain diseases.
Public access to the BigBrain dataset is being provided at no cost, with free registration at the CBRAIN portal.