Key result
A proposed cloud-based seizure alert system using EEG signals and Gaussian process classification was tested on the Amazon EC2 cloud for execution time and accuracy.
A proposed cloud-based framework using EEG signals and Gaussian process classification aims to automatically predict seizures in epileptic patients.
Cloud-based EEG seizure prediction feasible in technical tests; leaves open clinical validation before any patient use.
Epilepsy is a chronic disease that causes seizures. Automatic detection of a seizure before its occurrence could protect patients from accidents and potentially save their lives. This article proposes a framework that automatically predicts seizures by analyzing EEG signals. The wireless sensor technology captures brain signals, and cloud-based services are used to collect and analyze EEG data from a patient's mobile phone. Features from the EEG signal are extracted by using the fast Walsh-Hadamard transform (FWHT), and a higher-order spectral analysis is applied to FWHT coefficients to reduce the feature set relevant to normal, preictal, and ictal states of seizure. The entropy-based features are classified using the Gaussian process classification algorithm. The performance of the proposed model is tested on the Amazon EC2 cloud and assessed in terms of execution time and accuracy.
No takes yet. Share an insight, caveat, or question.
Sareen et al. (2016) studied Epilepsy. Cloud-based seizure alert system was evaluated on Execution time and accuracy. A proposed cloud-based seizure alert system using EEG signals and Gaussian process classification was tested on the Amazon EC2 cloud for execution time and accuracy.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: