AI uncovers benefits of mindfulness

20 Aug, 2019
 
AUT scientists Dr Zohreh Doborjeh and Dr Maryam Doborjeh
AUT scientists Dr Zohreh Doborjeh and Dr Maryam Doborjeh have discovered a link between mindfulness training and improved cognitive function.

Scientists at AUT’s Knowledge Engineering and Discovery Research Institute (KEDRI) have discovered a link between mindfulness training and improved cognitive function.

The research, published in Nature Scientific Report in April 2019, looked at the brain data for 40 participants before and after a six-week programme of mindfulness training.

A cross-university collaboration between KEDRI, AUT’s Department of Psychology and Nottingham Trent University in the United Kingdom, the research found that mindfulness training works for a range of people (with different levels of depression) in improving their cognitive function but is most successful for those with minor signs of depression.

KEDRI’s Research fellow Dr Zohreh Doborjeh says the initial cognitive testing helped them identify whether the participants were non-depressed, mildly depressed or severely depressed.

“We could see in the results that the participants with minor signs of depression had the greatest response to the mindfulness training. They showed increased cognitive function, such as motor planning, emotional expression, attention and emotional memory, which means they were able to focus more on their work and performed better in day to day tasks.”

The research developed sophisticated Spatio-Temporal Brain Data analytical tools based on one of the most promising trends of Artificial Neural Networks, called Spiking Neural Networks (SNN). “The deep brain-inspired SNN models could incrementally learn from brain dynamics gathered over time, in a 3D space and captured meaningful patterns from brain data.” says Dr Maryam Doborjeh, computer science lecturer at AUT.

This research was used to classify depressed participants who had a better response to mindfulness intervention compared to those in whom mindfulness training was less effective. “Through modelling the brain data with SNN, we were able to identify which areas of the brain responded more to treatment and which were more involved in improved function and how did it happen over time after the intervention” says Dr Zohreh Doborjeh.

“Mental health is an issue of huge concern in New Zealand, as we all know. Unfortunately, we can’t currently accurately predict which treatment will work for each patient so many patients face trying multiple different treatments before arriving at what works best for them. We hope this research can be further developed to help identify optimal, individualised treatment plans for people with depression, helping decrease the time taken in identifying the best course of treatment,” concludes Dr Zohreh Doborjeh.