Machine learning diagnoses depression with 75% accuracy

A cognitive neuroscientist has developed a supercomputer, trained with machine learning, to diagnose depression with 75 percent accuracy.

Using the Stampede supercomputer at the Texas Advanced Computing Center (TACC), David Schnyer, Phd, a professor of psychology at the University of Texas (UT) at Austin, trained the computer with a machine learning algorithm to identify common factors in patients with depression using MRI brain scans and genomics data.

"One difficulty with that work is that it's primarily descriptive. The brain networks may appear to differ between two groups, but it doesn't tell us about what patterns actually predict which group you will fall into," said Schnyer. "We're looking for diagnostic measures that are predictive for outcomes like vulnerability to depression or dementia."

Schnyer began by training the Support Vector Machine Learning system with examples of healthy and depression individuals. The computer was able to scan the data, identify points of depression and build a model for differentiating between healthy and depressed patients.

In a group of 45 healthy and 52 depressed participants, the machine learning system was able to analyze diffusion tensor imaging MRI scans. This brain data was analyzed by the Stampede computer to diagnose major depressive disorder with 75 percent accuracy.  

"Not only are were learning that we can classify depressed versus non-depressed people using DTI data, we are also learning something about how depression is represented within the brain," said Christopher Beevers, PhD, a professor of psychology and director of the Institute for Mental Health Research at UT. "One of the benefits of machine learning, compared to more traditional approaches, is that it should increase the likelihood that what we observe in our study will apply to new and independent datasets. That is, it should generalize to new data. This is a critical question that we are really excited to test in future studies."