PHINN-EEG Uses Topological Betti Curves to Classify Dream Content From EEG

arXiv·low signal

PHINN-EEG applies topological time-series analysis — dynamic Betti curves — to dream-state EEG for dream-content classification and topology-conditioned neural signal synthesis, moving beyond the power-spectral-density and statistical-moment features that cap prior work near 0.70 AUC on the DREAM database. It is a niche but genuinely novel application of topological data analysis to neural signal processing. Of interest to practitioners exploring TDA features for messy biosignal time series.