In a study in the Public Library of Science (PLoS) ONE journal, British researchers said their findings suggest it may be possible to design programs to predict which at-risk adolescents will go on to have psychiatric problems, giving doctors more time to intervene before illnesses set in.
"Combining machine learning and neuroimaging, we have a technique which shows enormous potential to help us identify which adolescents are at true risk of developing anxiety and mood disorders, especially where there is limited clinical or genetic information," said Janaina Mourao-Miranda of University College London, who led the study.
Depression and other psychiatric disorders are a major cause of death, disability and economic burden worldwide. The World Health Organization predicts that by 2020, depression alone will be the second leading contributor to the global burden of disease across all ages.
Two studies published late last year found that up to 40 percent of Europeans suffer from mental and neurological illnesses each year, and the annual cost of brain disorders is almost 800 billion euros.
But experts think that being able to diagnose potential problems earlier, and intervene to help at-risk young people, could significantly reduce the damage caused by psychiatric disorders and help ward off serious or recurrent illness.
As yet there are no known biological measures, or biomarkers, that can predict future psychiatric disorders and even genetic screening can't accurately predict individual psychiatric risk, the researchers explained in their study.
A family history of bipolar disorder, for example, confers a 10 percent risk of future bipolar disorder, but also a 10 to 25 percent risk of conditions like attention deficit hyperactivity disorder (ADHD), major depression and anxiety disorders and it is impossible to say which, if any, is more likely.
Mourao-Miranda's team took 16 healthy adolescents who each had a parent with bipolar disorder, and 16 whose parents had no history of psychiatric illness, and scanned their brains with a functional magnetic resonance imaging (fMRI) scanner while they performed a specially designed emotional test.
The researchers then used a computer program capable of machine learning to predict the probability that an individual belonged to either the low-risk or at-risk group and found it was accurate in three out of four cases.
The researchers also found the computer program predicted significantly higher risk probabilities for young people who were found in follow-up to have developed psychiatric disorders than for those who remained healthy at follow-up.
Mary Phillips of Pittsburgh University in the United States, who also worked on the study, said since most mental disorders begin in adolescence or early adulthood, early detection and treatment could delay or even prevent future illness.
"Anxiety and mood disorders can have a devastating effect," she said in a statement. "If we are able to identify those individuals at greatest risk early-on, we can offer early and appropriate interventions to delay, or even prevent, onset of these terrible conditions."
(Reporting by Kate Kelland, editing by Paul Casciato)