Deep learning enhanced seizure monitoring from wearable sensors

Deep learning enhanced seizure monitoring from wearable sensors Overmillion patients in the United States have uncontrolled epilepsy despite ongoing medical therapyWhen seizures are prolonged or violentthere is significant risk of injury or even Sudden Unexpected Death of EpilepsySUDEPwhich occurs in approximatelyinpatients per yearNovel seizure monitoring could help alleviate this burdenOur research team has created a software applicationEpiWatchto capture different sensor measurements related to seizure activity such as convulsionsaccelerometersheart rate increasesphotoplethysmography PPGand unresponsiveness to behavioral promptinginteractive user interfaceOur hope is to offer accurate seizure detection with improved false positive performance to encourage usageOur team proposes to develop multi modal sensor analysis driven by deep learning technology to enhance seizure monitoringWe are uniquely positioned to accelerate development by leveraging our team s prior EpiWatch IRB approved study which generated overhours of sensor dataIn Phase Iwe will teach EpiWatch how to read time series sensor data and how to discriminate seizure activityEpiWatch will employ a convolutional neural networka technique rooted in deep learningto self characterize seizure features from labeled sensor dataIn order to infer additional information from vast amounts of unlabeled sensor data from US epilepsy patientsEpiWatch will incorporate a deconvolutional neural network technique to increase predictive performanceA pilot study of EpiWatch will test its ability to identify the presence of seizures in a prospective new cohort of epilepsy patientsIf we are successfulwe envision a Phase II proposal which is focused on clinical translation of the technology and assessment of its impactOur goal is to combine recent advances in deep learning and scalable parallel computing to create EpiWatchIn the long termwe hope this monitoring technology will aid epilepsy patient management and improve outcomes PROJECT NARRATIVE Recurring seizures are disablingdangerousand often limit independenceWe are developing novel seizure detection using a consumer friendly device with wearable sensorsEpiWatchto enable monitoring and emergency alerting for seizures that occur without warningof all seizuresand without witnessesespecially when they are prolongedandgtminor accompanied by cardiac arrhythmias responsible for SUDEPSudden Unexpected Death with Epilepsyainannual risk for patients with uncontrolled seizuresIf emergency care can be summoned under these circumstancespatients can live more safely and independentlyin turn encouraging app usageWhen integrated with current disease monitoring activitiesEpiWatch will have the long range impact of providing a unique platform for individualized epilepsy care