Using technology to diagnose PTSD and its subtypes

Share this:

An HRI Research Trainee is using neuroimaging and machine learning to discover how post-traumatic stress disorder (PTSD) is related to patterns of brain activity, and the findings could have important implications for both diagnostic and preventative medicine.

Dr. Andrew NicholsonDr. Andrew Nicholson is a post-doctoral fellow affiliated with Western University, McMaster University and Lawson Health Research Institute. His postdoctoral fellowship is funded jointly by Mitacs Elevate and HRI, and his research focuses on the use of neuroimaging techniques to advance healthcare.

In a recent publication co-authored by HRI Associate Clinical Scientist, Dr. Ruth Lanius, and Homewood Research Chair in Mental Health and Trauma, Dr. Margaret McKinnon, Dr. Nicholson reported findings that suggest technological advances may enable early diagnosis of PTSD and its clinical subtypes.

About the study

Machine learning multivariate pattern analysis predicts classification of posttraumatic stress disorder and its dissociative subtype: a multimodal neuroimaging approach was published last month in Psychological Medicine.

In his study, Nicholson and his team used functional MRIs to document patterns of resting brain activity in three groups:

  1. Individuals with no history of PTSD (healthy control group)
  2. Individuals diagnosed with PTSD, and
  3. Individuals diagnosed with the dissociative subtype of PTSD (PTSD + DS)

The dissociative subtype of PTSD is characterized by symptoms of detachment and emotional numbness. It is different from the more familiar type of PTSD, which often results in difficulty controlling strong emotions and outbursts.

Nicholson’s study found that when patterns of brain activation were fed into a machine learning algorithm, the computer system predicted the classification of PTSD, PTSD + DS ,and healthy controls in new subjects with 91.63% accuracy.

Functional MRI images from Dr. Nicholson’s study show patterns of brain activity that are used to detect and classify a PTSD diagnosis.

How can these findings advance healthcare?

Dr. Nicholson’s findings suggest that distinct patterns of brain activity are associated with specific forms of PTSD. These brain activity patterns are considered unique biomarkers that may aid in the early diagnosis and intervention of PTSD and its subtypes. Nicholson’s findings may also help people with PTSD understand that there is a physical basis for their disorder.

New technologies can help us understand how symptoms of mental illness are related to brain activity. They enable the potential for detecting diagnoses that may not have been previously suspected or considered, and they facilitate expedient and objective diagnoses, opening the door to more individualized treatment approaches.

Dr. Nicholson’s work is unearthing valuable neurobiological clues about how to classify and better treat mental illness. To follow his research, visit drandrewnicholson.com.