Vwani Roychowdhury

The paper introduces a self-exciting population model to analyze calcium-imaging data from human brain assembloids, revealing increased history-dependent amplification in MAPT mutants.

The paper develops a time-resolved network analysis pipeline to detect differences in neuronal network dynamics between disease and control assembloids, finding increased degree variance and clustering in mutant networks.

This paper develops a transformer model to decode specific memories from human neuronal activity before and after sleep, finding differential contributions from brain regions.

PyHFO is a lightweight deep learning platform for efficient HFO detection in EEG data, achieving 50x speed improvements over standard methods.

The paper decodes human identity from neuronal activity using computer vision and deep learning, finding that tens of neurons in brain regions beyond MTL can robustly represent concepts.

This paper develops a deep learning algorithm to detect and classify epileptogenic high-frequency oscillations (eHFOs) from intracranial EEG data with high accuracy.