Fusion techniques of time frequency-based images to predict the outcome of rTMS depression therapy

The study introduces two fusion techniques—montage and blending—to combine EEG-derived time‑frequency images for predicting response to repetitive Transcranial Magnetic Stimulation (rTMS) in depression. Using a lightweight custom CNN, the authors evaluated the methods on two datasets (15 and 46 patients) under both segment‑level and subject‑disjoint cross‑validation. While segment‑level validation yielded high accuracies (up to 99.90 % on the primary dataset), subject‑disjoint validation showed poor performance, with AUC values ranging from 0.31 to 0.54 and a best subject‑level result of 0.874 ± 0.183 AUC.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.