The Design and Implementation of a Low-Cost Electroencephalogram to Predict Neural Disorders

We designed and developed an EEG to help predict neural disorders. This project achieved both high accuracy and low cost with cost-effective components such as the AD622ANZ instrumentation amplifier and TL048x operational amplifier. Signal processing software and hardware filters were implemented. In addition, a machine learning approach was utilized to develop a binary seizure classifier. Seizure noise was simulated, and future work would revolve around collecting live data from epileptic patients. Future work could show that our design could detect and diagnose other neural disorders. We aim to make this design a closed loop system and BCI (Brain-Computer-Interface) compatible.