Synap Sense

Electronics Engineering

Effective This project presents a real-time brain-monitoring system designed to detect epileptic seizures and loss of consciousness (LOC) using a low-cost, single-channel EEG sensor interfaced with a Raspberry Pi. The system aims to provide an affordable and portable solution for continuous brain-state monitoring, especially in home-care or remote medical settings where immediate medical intervention is crucial. EEG signals are acquired from the prefrontal cortex using a three-electrode frontal lobe sensor. These analog signals are digitized and wirelessly transmitted to a Raspberry Pi via Bluetooth or UART. The data is processed in 10-second windows through baseline correction, z-score standardization, and optional noise filtering to prepare for real-time feature extraction. Time-domain, frequency-domain, and non-linear features are computed from each segment to characterize brain activity. A pre-trained Random Forest classifier, optimized using offline data, then categorizes each signal segment as either normal, seizure, or LOC. Upon detection of abnormal brain states, the system automatically activates a SIM900A GSM module to alert caregivers—sending an SMS for LOC detection and both an SMS and a voice call for seizure events. The entire pipeline is designed to run in real-time with minimal latency, offering a practical solution for medical alert systems, especially in underserved regions.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.