Using Machine Learning to Diagnose Depression
Machine learning is endemic is spreading to more and more areas. It's making a splash in healthcare, even diagnosing hard-to-identify disorders.
Join the DZone community and get the full member experience.Join For Free
Machine learning has been increasingly capable of accurately diagnosing a range of physical and mental health conditions in recent years. I’ve written previously about algorithms that monitor things like speech to detect the onset of conditions such as Alzheimer’s, whilst applications have also used mobile phone data to detect changes in lifestyle and possible depression in individuals.
A recent study from the University of Texas at Austin takes a slightly different tact by using AI to spot vulnerability to depression from brain imaging.
The researchers worked with a supercomputer to train the algorithm to detect commonalities in MRI scans, genomic data and various other datasets relevant to depression and anxiety. It aims to improve upon previous work by researchers who have studied mental disorders via the relationship between brain function and structure in neuroimaging 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,” the researchers say. “We’re looking for diagnostic measures that are predictive for outcomes like vulnerability to depression or dementia.”
The team utilized Support Vector Machine Learning to develop their prototype, which was able of correctly classifying individuals with a major depressive disorder with 75% accuracy.
The system was trained with data from nearly 100 patients, with an even split of those already diagnosed with depression, and a healthy control group. Each participant received diffusion tensor imaging (DTI) MRI scans. These are designed to tag water molecules to determine the extent to which those molecules are microscopically diffused in the brain over time.
The approach appeared to be effective, as it proved able to accurately classify people according to their mental health. What’s more, the predictive information was also distributed across the brain rather than being localized.
“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,” the researchers say. “Rather than trying to find the area that is disrupted in depression, we are learning that alterations across a number of networks contribute to the classification of depression.”
Working at Scale
The authors believe that the complexity of the brain makes machine learning essential, as detecting the relationship between the 175,000 or so voxels in the brain is impossible for humans.
Suffice to say, whilst the initial results are promising, more work needs to be done before this approach can be used in a clinical setting. The aim is to utilize more data, both in terms of MRI scans and also genomic and other data to better train the system.
“One of the benefits of machine learning, compared to more traditional approaches, is that machine learning 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,” the team say. “This is a critical question that we are really excited to test in future studies.”
Published at DZone with permission of Adi Gaskell, DZone MVB. See the original article here.
Opinions expressed by DZone contributors are their own.