EPILEPTIC SLEEP SEIZURE DETECTION AND NOTIFICATION SYSTEM

Abstract

Epilepsy is a disease that affects around 50 million people around the world, and around 3.4 million adults in the US present active epilepsy, meaning they still suffer from seizures due to this condition. All these patients are at risk of Sudden Unexpected Death in Epilepsy (SUDEP), which is most likely to happen during sleep, as these seizures often go unnoticed by caretakers. There are a series of devices currently in the market ranging from bracelets to mattress pressure sensors that try to detect seizures, but often are inaccurate presenting many false positives, having a negative impact in the quality of sleep of caregivers and patients. Here, an open-source, minimally invasive system is presented that utilizes stereoscopic IR cameras along with several post-processing techniques utilizing Fast Fourier Transforms and artificial intelligence to detect seizures with better accuracy focused on returning some of the lost quality of life of the patients.