Researchers at MIT’s Computer Science and Artificial Intelligence Laboratory have devised a new way to combine information from two different kinds of MRI scans, allowing drug developers to map which regions of the brain are affected by certain neurological disorders.
Neurological diseases, such as schizophrenia, are known to affect certain regions of the brain. Likewise, most drugs made to treat such diseases are made to act upon certain regions. The complication comes because those regions connect to other regions in way that no single brain scan can easily show. Polina Golland, an associate professor of computer science at MIT, has come up with an algorithm, or formula, which uses the information on the connections between them to map those regions. The algorithm is applied to two types of MRIs, both of which have been in use for at least 20 years: Diffusion MRIs, which look at the substance of the brain, as compared to functional MRIs, which look at brain activity. Golland said both types of scans are “very noisy,” and traditional analysis of them has focused on comparing hundreds of scans from different patients, rather than on extracting more information from different views of a single brain.. Click on the image or title to learn more.