Dr. Joshua Curtiss, Assistant Professor at Northeastern University, has published groundbreaking research in Clinical Psychology Review that could transform how clinicians predict treatment outcomes for patients with anxiety and depression.
The comprehensive meta-analysis, titled “Machine learning in the prediction of treatment response for emotional disorders: A systematic review and meta-analysis,” examined 155 studies to evaluate the effectiveness of machine learning algorithms in predicting treatment response for patients with emotional disorders. Co-authored with Christopher DiPietro, the study analyzed data from 3,816 records to assess machine learning’s prediction accuracy across evidence-based psychotherapies, pharmacotherapies, and other treatments for emotional disorders.
The research found that machine learning methods achieved an overall mean prediction accuracy of 76%, with neuroimaging data proving to be the most effective predictor compared to clinical and demographic data. The results also suggest that machine learning models had greater difficulty predicting treatment response for depressed patients as opposed to other populations (e.g., anxiety disorders). The findings suggest that machine learning could significantly reduce the trial-and-error approach in mental health care, helping patients receive effective treatment more quickly.
The study was published in Clinical Psychology Review, a leading journal in clinical psychology with an impact factor of 13.7, highlighting the significance of this research contribution to precision medicine approaches in mental health treatment.
