Detecting hypermotor seizures using extreme value statistics

Abstract

Nocturnal home monitoring of epileptic children is often not feasible due to the cumbersome manner of seizure detection with the standard method of video/EEG-monitoring. We propose a method for hypermotor seizure detection based on accelerators attached to the extremities. Hypermotor seizures often involve violent movements with the arms or legs, which increases the need for an alarm system as the patient can injure himself during the seizure. In the literature, classification models are commonly estimated in a supervised manner. Such models are estimated using annotated examples. This annotation of data requires expert (human) interaction and results therefore in a substantial cost in the estimation process of the seizure detection model. In this work we propose the use of an unsupervised approach for estimating seizure detection models. Our method does not require any annotation of data while obtaining state-of-the-art classification scores that are comparable to those of a model estimated in a supervised manner. The proposed methodology is based on extreme value statistics. Since epileptic seizures are rare compared to normal movements, all recorded data can be used to estimate a model of normality. The very few abnormal events have a negligible effect. As a consequence, a person-dependent epileptic seizure detector can be estimated with little human interaction. After segmenting the acquired acceleration signals in movement events, features are extracted for further processing. On the basis of the preprocessed data a classification model for discriminating epileptic and non-epileptic movement events is estimated. Using this approach we were able to detect hypermotor seizures in two patients with a sensitivity of 100.00% and a PPV of 85.4% for one patient, and 98.8% and 79.1% for the sensitivity and PPV respectively for the other patient. We also found that an increase of the number of features did not increase the performance substantially.