The AI doctor is in: Mental health analysis may soon be done by robots

By Joe Morgan via SWNS

Mental health assessments could soon be done by robots, according to a new study.

Instead of a one-on-one phone or in-person conversation with a licensed professional, artificial intelligence (AI) could soon be allowed to search through clients' social media or mobile phone data to produce 'personalized' assessments.

By tracking behavioural patterns, biological and medical data, a new model was built that could carry out the classical testing for mental health procedures.

But they insist that human mental health practitioners will still need to be able to be involved to check, verify and interpret the data that is discovered.

The World Health Organisation (WHO) has determined there has been a 13 percent increase in mental health conditions and substance abuse disorders between 2007 and 2017.

In the pandemic, many National Health Service clinics in the United Kingdom have said their mental health teams are overwhelmed by many people who sought help for anxiety and depression.

Researchers say that by building a computer model that can question and understand personal circumstances and habits, the AI could discover strong correlations with specific diseases or outcomes and also save medical professionals time and work.

The researchers took advantage of the UK Biobank, one of the world's largest and most comprehensive biomedical databases, that contains detailed and secure health-related data on the UK population.

The scientists built models that approximate measures for brain age, and scientifically defined intelligence and neuroticism traits.

Developing approximations in this way has been successfully employed in the past for predicting “brain age” from MR images.

Dr. Denis Engemann, a research scientist at Inria Saclay Research Centre in France, said: “In this work, we generalized this methodology in two ways.

"First, we demonstrated that, beyond biological aging, the same proxy-measure framework is applicable to constructs more directly related to mental health," Engemann said. "Second, we showed that useful proxy measures can be derived from other inputs than brain images, such as sociodemographic and behavioural data.”

He added: “What is not going to change is that mental health practitioners will need to carefully interpret and contextualise test results on a case-by-case basis and through social interaction, whether they are obtained using machine learning or classical testing.”

The study was published in the journal GigaScience.