Application of Low-cost Mobile Health for Remote Monitoring of Epilepsy Patients (Preprint)

UNSTRUCTURED The objective of this study is to investigate the feasibility of smartphones for processing larger electroencephalography (EEG) recordings for the application towards remote monitoring of epilepsy patients. We have developed a mobile application to automatically analyze and perform the classification of epileptic seizures. For this purpose, we have used the cross-database model developed in our previous study using successive decomposition index and matrix determinant as features, adaptive median feature baseline correction to overcome inter-database feature variation and post-processing based support vector machine for classification using five different EEG databases. The sezect (seizure detect) Android application was built using Chaquopy soft- ware development kit which uses Python language in Android Studio. Different duration of EEG signals was tested on different versions of smartphones using sezect app to check its feasibility. The computational time required to process the real-time EEG data on smartphone and classification results suggests that mobile-health could be a great asset to monitor epilepsy patients. More details on sezect Android app can be found at: http://doi.org/10.5281/zenodo.3592415.