A newly developed artificial intelligence (AI) system can now detect depression and anxiety by analysing a child's speech pattern.
One in five children reportedly suffers from depression and anxiety. Because children under eight-years-old are not able to clearly explain their emotional struggles, it is difficult for adults to recognise the potential mental health issue. This new technique can help doctors diagnose difficult conditions, like these mental health issues, at a much early stage.
"We need quick, objective tests to catch kids when they are suffering," study author Ellen McGinnis, a clinical psychologist at the University of Vermont Medical Center's Vermont Center for Children, Youth and Families, told a news portal. Adding, "The majority of kids under eight are undiagnosed."
Researchers say early diagnosis is crucial because children have a more positive response to treatment while their brain is still in the developmental stage. Substance abuse and suicide are some of the big risk factors that children may be more prone to if these health issues are not addressed early.
For the study, researchers wanted to find a way to make the process for diagnoses more efficient and faster. So they decided to used a mood induction task known as the Trier-Social Stress Task. Its primary intention is to induce feelings of stress in subjects. A clinical interview and parent questionnaire were also used to better diagnose the children who were part of the study.
The 71 children, who took part in the research were between the ages of three and eight. They had to come up with a story and talk about it for three minutes. Researchers explained to the children that they would be judged on how interesting their stories were. The team's feedback was either neutral or negative. "The task is designed to be stressful and to put them in the mindset that someone was judging them," Ellen McGinnis told a news portal.
The team then used the system to examine audio recordings of every child's story. Researchers discovered the algorithm was extremely helpful in diagnosing children. "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," senior study author Ryan McGinnis.
Speaking about how the system works, Ellen McGinnis told a news portal: "A low-pitched voice and repeatable speech elements mirrors what we think about when we think about depression: speaking in a monotone voice, repeating what you're saying,"
Now, the team wants to develop a speech analysis algorithm for clinical use. The study's findings were originally published in the Journal of Biomedical and Health Informatics.