Researchers at the Mayo Clinic have made the first step toward applying artificial intelligence (AI) to predict early outcomes with antidepressants in children and adolescents with major depressive disorder. They discovered variations in six depression symptoms: trouble having fun, social withdrawal, excess fatigue, irritability, low self-esteem, and depressed sentiments. The study is published in The Journal of Child Psychology and Psychiatry.
Over 50 per cent of all mental health disorders are diagnosed when individuals are young (under the age of 18). As a result, it is critical to establish predictive approaches for treatment outcomes in depressed youth, according to Arjun P. Athreya, PH. D, M.S, a Mayo Clinic researcher and lead author of the study. This research addresses a global public health problem as paediatric depression is common and often undertreated.
The ‘Children's Depression Rating Scale, Revised’ was used by the researchers to assess these symptoms and predict the results of 10 to 12 weeks of an antidepressant pharmacotherapy programme.
In fluoxetine testing datasets, the six symptoms predicted 10 to 12-week results with an average accuracy of 73 per cent at four to six weeks, whereas in duloxetine testing datasets, the same six symptoms predicted 10 to 12-week results with an average accuracy of 76 per cent at four to six weeks.
Predicting response and remission accuracy in placebo-treated patients was much lower than in antidepressant-treated individuals, at 67 per cent.
“This preliminary work suggests that AI has promise for assisting clinical decisions by informing physicians on the selection, use, and dosing of antidepressants for children and adolescents with major depressive disorder," says senior author and Mayo Clinic Child Psychiatrist Paul Croarkin, D.O. "We saw improved predictions of treatment outcomes in samples of children and adolescents across two classes of antidepressants."
The current findings demonstrate that an AI platform using machine learning can predict antidepressant treatment outcomes in children and adolescents early during therapy. This is highly beneficial in the future as it has the potential to assist busy clinicians in treatment planning for adolescents while avoiding exposure to treatments that may not be effective for a patient. Furthermore, the findings show the potential of AI and patient data to ensure that children and adolescents receive treatment that has the highest likelihood of delivering therapeutic benefits while minimising side effects, according to Dr Athreya.
"We designed the algorithm to mimic a clinician's logic of treatment management at an interim time point based on their estimated guess of whether a patient will likely or not benefit from pharmacotherapy at the current dose," says Dr Athreya. "Hence, it was essential for me as a computer engineer to embed and observe the practice closely to not only understand the needs of the patient, but also how AI can be consumed and useful to the clinician to benefit the patient."
This integration makes use of clinical symptom changes that occur early during treatment. Patients undergo standard interviews, but using this platform, they may be given accurate predictions of treatment outcomes.
The findings of the study serve as a foundation for future research incorporating physiological information, brain-based measures, and pharmacogenomic data for precision medicine approaches in treating depressed youth. This will improve the care of young children suffering from depression and assist physicians in initiating and dosing antidepressants in patients who will benefit the most.
"Technological advances are understudied tools that could enhance treatment approaches," says Liewei Wang, M.D., PhD, the Bernard and Edith Waterman Director of the Pharmacogenomics Program and Director of the Centre for Individualised Medicine at the Mayo Clinic. "Predicting outcomes in children and adolescents treated for depression is critical in managing what could become a lifelong disease burden."
The study was a collaborative effort between the Mayo Clinic departments of Molecular Pharmacology and Experimental Therapeutics, as well as Psychiatry and Psychology, with assistance from the Mayo Clinic Centre for Individualised Medicine.
This present research is promising, although it is still in its early stages. Dr Athreya adds that future research will integrate biological data or "biomarkers" to enhance or increase diagnostic accuracy and treatment planning.
