While thorough examination of the affected organ is standard practice in many medical fields, psychiatric treatment decisions remain largely guided by expert opinion and patient history. Over the past decade, however, mounting evidence has shown that electroencephalogram (EEG) and electrocardiogram (ECG) time series data contain valuable information that can not only enhance diagnostic processes—moving beyond rigid diagnostic categories—but, more importantly, assist in matching patients with the most effective treatments.

This talk will explore the development of electrophysiological treatment markers, from conventional research findings such as frontal alpha asymmetry and EEG vigilance, to cutting-edge advancements in predictive EEG research using deep learning models. These innovations significantly improve the accuracy of treatment predictions. The presentation will highlight results from large international datasets on depression and other psychiatric disorders, offering insights into the future of personalized psychiatry. Further, an outlook will be given on the usage of automated analysis pipelines for clinical usage and as software as medical devices.