Neurospectrum: A Geometric and Topological Deep Learning Framework for Uncovering Spatiotemporal Signatures in Neural Activity

Neurospectrum is a deep learning framework designed to analyze neural activity by encoding it as latent trajectories that reflect spatial and temporal structures. It uses graph-based representations and a manifold-regularized autoencoder to preserve temporal geometry, extracting features like curvature and persistent homology. Neurospectrum was tested on both simulated and experimental datasets, demonstrating its ability to track phase synchronization, reconstruct visual stimuli, and identify biomarkers of obsessive-compulsive disorder. The framework outperforms traditional methods in uncovering meaningful neural dynamics, offering a modular and interpretable approach to neural signal analysis.