A machine-learning model used pretreatment symptom scores and electroencephalographic (EEG) measures to reliably predict patients with depression who responded to antidepressant treatment, according to a study published online in JAMA Network Open.
“In this study, we developed a machine learning algorithm, ElecTreeScore, to evaluate the association of objective EEG measures acquired before treatment with the prediction of acute antidepressant response for individual symptoms of depression,” wrote researchers from Stanford University, California, Harvard Medical School, Boston, Massachusetts, and Imperial College London.
“Under this approach, we took into account the important associations between baseline symptom severity and treatment-associated change in symptoms and considered the association of EEG features in their own right and to what extent EEG features have a meaningful association with outcomes in addition to symptom severity.”
The prognostic study included 518 adult outpatients with major depressive disorder.
Using gradient-boosted decision trees, the machine-learning model demonstrated high discriminative performance for identifying improvement in specific symptoms, researchers reported. The score of the symptom at baseline proved the most important feature in predicting symptom improvements. Although smaller, associations between EEG features and prediction of specific symptom improvements were also meaningful.
“Overall,” researchers concluded, “our findings build on prior work in 2 key ways: first, by demonstrating that predictive models can capitalize on established roles for using EEG markers to quantify neural activity in psychiatric illness to predict treatment-associated changes over time, and second, by explicitly using individual symptoms as independent outcome variables, to parse the extreme heterogeneity of major depression.”
Future studies should validate the model’s performance prospectively and in independent samples and clinical settings, they advised.
Rajpurkar P, Yang J, Dass N, et al. Evaluation of a machine learning model based on pretreatment symptoms and electroencephalographic features to predict outcomes of antidepressant treatment in adults with depression: a prespecified secondary analysis of a randomized clinical trial. JAMA Network Open. 2020;3(6):e206653.