Epilepsy is a neurological disorder affecting millions worldwide, with a significant burden in low-resource settings where diagnostic tools are often inaccessible. This project aims to develop a low-cost, edge-device-compatible EEG seizure detection system tailored for rural communities. The goal is to leverage publicly available EEG datasets for model development and subsequently collaborate with local institutions for dataset expansion and real-world validation in Bangladesh.

Data Sources
Public EEG Datasets: The CHB-MIT Scalp EEG dataset will be used for initial model training and validation. This dataset provides extensive EEG recordings from pediatric epilepsy patients.
Local Data Collection (Future Scope): Collaboration with the National Institute of Neuroscience (NINS), Bangladesh, will facilitate access to region-specific EEG recordings, ensuring the model’s robustness and generalizability.