EEG Test Predicts Which Depression Patients Will Respond to Magnetic Brain Stimulation

Summary

A team at Anhui Medical University led by Li Wan built a biomarker, the dlPFC sensitivity index, that uses EEG recorded while patients perform an emotional attention task to forecast how well they will respond to repetitive transcranial magnetic stimulation (rTMS) for major depression. The biomarker is calculated from how strongly connectivity between the prefrontal cortex and other emotion-processing regions shifts under conflicting emotional cues, and an AI model built on this measure sorted patients into likely responders with 96% area-under-curve accuracy in an initial group of 116 people. In a follow-up trial of 67 patients randomized to active or sham rTMS, those flagged as "sensitive" beforehand showed significantly larger improvements in attention-task performance and reductions in hostility and paranoia symptoms than both non-sensitive and sham patients, with EEG connectivity changes afterward confirming the predicted brain circuits had actually shifted.