Lifestyle

How AI could help speed up diagnoses of anxiety and depression in children



Artificial intelligence (AI) can do a lot of things, such as create poetry, help you understand bias, or discover the next hit start-up.

Now Academics at the University of Vermont have created a machine learning algorithm that can detect signs of anxiety and depression in the speech patterns of young children, as a way to speed up the accuracy of diagnosing conditions. 

Using AI to spot signs in speech is important because often children under the age of eight aren’t able to articulate how they are feeling. However, early diagnosis is crucial because children will respond well to treatment while their brains are still developing. If left untreated, it may mean they are at greater risk of substance abuse and suicide later in life. 

Lead author of the study, Ellen McGinnis, clinical psychologist at the University of Vermont Medical Center’s Vermont Center for Children, Youth and Families, said: “We need quick, objective tests to catch kids when they are suffering. The majority of kids under eight are undiagnosed.”  

To test out the AI, a group of 71 children aged between three and eight underwent a typical structured clinical interview and parent questionnaire – the usual ways of identifying anxiety and depression in children.

In addition, they were asked to improvise a three-minute story, and told they would be judged based on how interesting it was. The task was designed to be stressful and to put them in the mindset of someone who was judging them, according to McGinnis. A buzzer would sound after 90 seconds and again with 30 seconds left to inform the children how much time they had left. 

Then, the researchers applied the algorithm to analyse features of the audio recordings and found that the algorithm was highly successful at diagnosing children, usually only taking a few seconds to process. In general, low-pitched voices with repeatable speech inflections and content, as well as a high-pitched response to the surprising buzzer, was able to demonstrate instances of anxiety and depression. 

“The algorithm was able to identify children with a diagnosis of an internalizing disorder with 80 per cent accuracy, and in most cases that compared really well to the accuracy of the parent checklist,” said Ryan McGinnis, another author on the study. 

The next step will be to develop a speech analysis algorithm that could be used for clinical use, such as via a smartphone app. In general, the aim is to help identify children at risk before their parents suspect anything is wrong. 

There are other ways that AI is being used to signal depression, such as one notable study from 2017 that used Instagram photos as predictive markers of depression. In the UK alone, around one in four people will experience a mental health problem every year, so finding faster ways to diagnose and treat the problem is in the best interests of everyone.  

 

Subscribe to Women Tech Charge, a new podcast from the Evening Standard.

From fashion to finance, technology is revolutionising our lives. Meet the extraordinary women who are leading the charge.

Listen on Apple Podcasts, Spotify or wherever you get your podcasts.



READ SOURCE

Leave a Reply

This website uses cookies. By continuing to use this site, you accept our use of cookies.