Predicting subject traits from brain spectral signatures: an application to brain ageing (Updated October 30, 2024).

The article discusses the use of electroencephalography (EEG) data to predict subject traits, particularly focusing on brain aging. The study explores the application of data-driven kernel methods for prediction from single EEG channels by reinterpreting the spectrogram as a probability distribution. The research suggests that these kernel methods show improved performance in predicting biological age, demonstrating the capacity to generalize across different experiments and acquisition setups. The preprint has not yet undergone peer review and provides valuable insights into the potential of EEG data in neuroscience and clinical research. [Extracted from the article]