Federal Project Grant Award Summary

Beth Israel Deaconess Medical Center, Inc. received a $252,060 Project Grant award from the National Institute of Neurological Disorders and Stroke (NINDS) under the Extramural Research Programs in the Neurosciences and Neurological Disorders program (CFDA 93.853). The award, effective May 1, 2026 through April 30, 2028, funds development of SourceSpikeNet, a biophysically grounded artificial intelligence system designed to improve detection and localization of interictal epileptiform discharges (IEDs) in electroencephalogram (EEG) recordings. Current clinical practice relies on subjective identification of IEDs with significant inter-reviewer variability and high false alarm rates in automated systems, creating a critical clinical need that this project addresses.

The research deliverables include three primary objectives: (1) characterization of IED prototype patterns from 32,433 expert-annotated discharges using inverse source modeling and self-supervised machine learning; (2) development of SourceSpikeNet through deep learning algorithm training on an augmented dataset of 3.2 million simulated spikes from 1,000 cortical locations; and (3) comprehensive evaluation of the system's performance on continuous EEG recordings from 2,484 patients over 3-7 day periods with comparative analysis of IED localization accuracy versus traditional methodology. Upon completion, the project is expected to deliver superior spatiotemporal localization capabilities with reduced false alarm rates and enhanced spatial resolution, potentially improving seizure focus identification for surgical planning in epilepsy patients.