Estimating Medication Status via Resting-State Functional Connectivity in Major Depression

This study aims at developing a multivariate pattern analysis (MVPA) approach for making accurate predictions about antidepressant medication status of major depressive individuals based on resting-state functional connectivity magnetic resonance imaging (rs-fcMRI) data. The experimental results showed that, by training the linear support vector machine classifier with principal component analysis on the data of 24 depressed patients and 29 demographically similar controls, the medication status of 16 patients who underwent treatment with the antidepressant medication and subsequently achieved clinical remission could be predicted successfully. Moreover, the most discriminating functional connectivities were located within or across the cerebellum, default-mode network, and affective network, indicating these networks may play important roles in major depression. This study not only demonstrated that MVPA methods can correctly predict medication status of depressed patients with clinical remission, but also suggested that rs-fcMRI can provide potential information for clinical diagnosis and treatment.