Applied Projects#
Self-contained projects using recorded EEG and observed participant metadata. Age, sex and p-factor projects select six ds005505 resting-state participants (approximately 595.8 MB of signal files) and evaluate at participant level. The clinical summary uses metadata only. P300 transfer uses recorded oddball events. The eyes-open/closed CNN project trains ShallowFBCSPNet within one ds005514 participant (about 90 MB) using block-wise train/test splits. Set EEGDASH_CACHE_DIR to reuse acquisitions. These small selected cohorts demonstrate analysis workflows, not clinical or population claims.
Predict observed age (years) from real EEG features
Exploratory eye-state decoding in an outcome-selected participant
Familiar vs unfamiliar faces with EEGPrep and ShallowFBCSPNet
P300 transfer between participants with MMD adaptation
Regress observed p-factor with a neural EEG decoder
