Psychologists and cognitive neuroscientists are currently studying the ability of the Stampede supercomputer in order to provide accurate predictions of risk for those with depression and anxiety. Stampede is one of the most powerful computing machines in the world. It is dedicated for open science research. the supercomputer was funded by the National Science Foundation Grant ACI-1134872, and developed with the companies Intel, Dell and Mellanox and operated from the Texas Advanced Computing Center (TACC) at the University of Texas at Austin, U.S.. Stampede, which has been operational since 2013, is composed of 6400 nodes, 102400 CPU cores, 205 TB total memory, 14 PB total and 1.6 PB local storage.
Using Stampede to help medics to review data to detect depression involved developing a machine learning algorithm. The program is intended to identify commonalities among patients using Magnetic Resonance Imaging (MRI) brain scans, genomics data and other factors in order to make predictions of risk for those with depression and anxiety. From the analysis of hundreds of patient data inputs (taken from 2 treatment-seeking participants with depression, and 45 healthy control participants), the researchers have successfully classified individuals with major depressive disorder with a 75 percent accuracy. This provides the basis for a workable diagnostic tool.
What is key to the success is ‘machine learning’. Machine learning, a type of artificial intelligence, involves the construction of algorithms that allow the machine to “learn” by constructing a model from sample data inputs. This allows the machine make independent predictions on new data, ideally becoming better over time.
READ MORE: Can probiotics reduce symptoms of depression?
For the next steps the research group aim to advance machine learning using Stampede 2. This version of the supercomputer comes online later on in 2017. The second generation will allow for more powerful computer processing. It is hoped this increased processing capacity will allow for even greater accuracy.
The research is published in the journal Psychiatry Research: Neuroimaging, under the heading “Evaluating the diagnostic utility of applying a machine learning algorithm to diffusion tensor MRI measures in individuals with major depressive disorder.”