Technology that predicts songs you may like on iTunes or items you might be interested in on Amazon.com, is more useful than just helping you find your next favorite band, it can also help find cures for mental illness. This technology, known as “machine learning”, is what companies like Amazon.com and Apple use to recommend content to their customers.
Terran Lane, associate computer science professor at the University of New Mexico, is using machine learning technology to teach a computer how to recognize how different networks in the brain communicate with one another.
“We’re looking at what regions of the brain are talking to each other,” he said. “We are comparing someone with mental illness to someone with a healthy brain and seeing if there is any difference in how parts of the brain are communicating. Hopefully, we can better understand mental disease so that doctors can provide better treatment.”
Lane is working closely with Vince Calhoun, Chief Technology Officer of the UNM Mind Research Network and electrical engineering professor. The two have also been working closely with researchers from the University of California, Irvine. Calhoun is using functional magnetic resonance imaging (FMRI) to take three-dimensional images of the human brain.
“We can take pictures of the brain every couple of seconds that gives us a 3-D image,” Calhoun said. “Over time we can see how oxygenation goes up in the brain. We’re looking at how blood flow in the brain is different from people that don’t have a mental illness. The long-term goal is to take that information and pull out the interesting pieces of it in patients with schizophrenia or bipolar disorder.”
An FMRI looks at where activity in the brain is happening. Test subjects are asked to do an activity such as think about a certain thing or look at an image. Elizabeth Browning, graduate student in the Neurocognitive Psychology Department at UNM and former employee at the Mind Research Network, said an FMRI is useful in detecting what parts of the brain are active.
“An FMRI shows where oxygen is flowing through the brain and looks at which networks in the brain are working during a task,” she said. “A schizophrenic’s brain will light up differently than a control person’s brain.”
Calhoun has scanned people at UNM, as well as the University of Minnesota, the University of Iowa and Massachusetts General Hospital.
There is hope this research can one day be used to treat patients, but for now Calhoun said it is still a long way away from being used for treatment.
“We’re not yet at the point where we’ve actually applied it in the clinic, but we have a lot of evidence that suggests that it should be useful,” he said.
Browning believes computer imaging is how doctors and psychologists will come to understand the brain in the future.
“This way of looking at the underlying structures of the brain in mental disease is the way of the future,” she said. “We’re getting closer and closer to mapping out every single network in the brain.”