Using artificial intelligence, a brain scan and medical data from a person suffering from depression, Dutch scientists were able to determine whether psychotropic medications would be effective.
- Until now, no tool could predict whether antidepressants would be effective. Six to eight weeks after taking the medications, doctors check to see if they have worked.
- Recently, Dutch researchers developed an algorithm, which successfully predicted that only a third of participants in a study would respond to treatment.
- According to the authors, blood flow in the anterior cingulate cortex, the brain area involved in emotion regulation, would be predictive of the effectiveness of antidepressants.
In depressive disorder, the response to antidepressants varies greatly from person to person, making the process of finding effective treatment longer. Recently, researchers from UMC Amsterdam and Radboudumc (Netherlands) wanted to know if they could predict the effects of sertraline, one of the most commonly prescribed psychotropic drugs in the United States and Europe. .
A third of depressed patients respond to antidepressant, according to AI
As part of their study, published in the journal American Journal of Psychiatry, the team recruited 229 adults suffering from major depression. Participants had to have clinical exams and MRI scans before and after receiving sertraline or a placebo for a week. Next, the scientists developed an algorithm to determine whether they could predict the effectiveness of the antidepressant.
The analysis showed that only a third of patients would respond to the drug. “This is important news for patients. Normally, it takes 6 to 8 weeks before knowing whether an antidepressant will work. (…) Thanks to this method, we can already avoid two thirds of the number of ‘erroneous’ prescriptions for sertraline and thus offer a better quality of care to the patient. Because the drug also has side effects”, has explained Liesbeth Renemanauthor of the work.
Brain: blood flow in the anterior cingulate cortex, the key to prediction?
According to Eric Ruhé, a psychiatrist at Radboudumc, the algorithm suggested that blood flow in the anterior cingulate cortex, namely the area of the brain involved in emotion regulation, would be predictive of the effectiveness of the drug. “And at the second measurement, a week after onset, the severity of their symptoms was found to be predictive.”
The authors believe that in the future, this simple and effective protocol could help to better adapt treatment to each patient. In the future, they plan to improve the algorithm by adding additional information.