image: ©Prostock-Studio | iStock

By analysing patient characteristics, clinical data, and treatment preferences, the new AI system, created by University College London, recommends antidepressants more likely to work for each individual, aiding doctors in selecting suitable medication.

Researchers say this technology could reduce the trial-and-error process patients face when starting antidepressants. Clinical trial results suggest that patients whose medication was selected using the AI system were less likely to stop treatment and showed greater improvements in depression symptoms.

The PETRUSHKA tool supports clinicians and patients in choosing an antidepressant tailored to each individual. The tool, developed by the University of Oxford and funded by an NIHR Research Professorship, is the first mental health clinical prediction tool to be found effective.

The tool uses AI to combine clinical and demographic information with patient preferences, including side effects. It takes three minutes to be administered and was tested in a randomised clinical trial across the UK, Brazil, and Canada. It involved more than 500 adults with major depressive disorder and launched in 2024.

The researchers showed that patients whose antidepressant was selected with the PETRUSHKA AI tool were significantly more likely to continue treatment and experience lasting symptom improvement than those in usual care.

In the trial, participants using the AI tool were around 40% less likely to discontinue their antidepressant within the first eight weeks of treatment. Fewer people stopped treatment because of adverse effects. By 24 weeks, those in the PETRUSHKA group also reported greater improvements in depressive and anxiety symptoms.

The tool was co-produced with people with lived experience of depression and has been designed to be used in everyday clinical settings. Clinicians and patients used PETRUSHKA together during a consultation.

NIHR Research Professor, Andrea Cipriani, Professor of Psychiatry at the University of Oxford, honorary consultant psychiatrist at Oxford Health NHS Foundation Trust and lead investigator of the study, said: “Mental health is lagging behind other fields of medicine and for too long, antidepressant treatment has relied on trial and error.

Mike Lewis, NIHR Scientific Director for Innovation, said: “This study shows the real power of combining digital technology and personalised treatment.

“By harnessing data and embracing cutting-edge digital tools, we can tailor care more precisely to each patient – improving outcomes for individuals. Forward-thinking research such as this shows the impact of using innovative technologies to ensure patients across the UK receive smarter, more effective care.”