[News Viscovery]
Munich (DE) / Melbourne (AU) / Vienna (AT), 6 February 2023
Together with Nicolas Rost and his colleagues from the Max Planck Institute of Psychiatry, we published the article Multimodal predictions of treatment outcome in major depression: A comparison of data-driven predictors with importance ratings by clinicians in the Journal of Affective Disorders.
The aim of this article is to find reliable models to predict treatment outcomes for patients with major depressive disorder. To this end, multiple outcome data were clustered using Viscovery´s SOM-Ward algorithm and with a consensus clustering approach utilizing a Python k-medoids algorithm. The resulting cluster models are in good accordance with each other and define useful outcome classes. In a second step, supervised machine learning methods, namely logistic regression and random forest, were used to predict outcome classes based on the patients´ baseline assessments.
Find the full article at sciencedirect.com