Comparison of research results is difficult as studies focus on different seizure types, timing (night versus day) and patients (adult versus pediatric patients). Nevertheless, we are convinced that the most effective seizure detection systems are multimodal, combining for example detection methods for movement and heart rate, and that devices should especially take into account the user's seizure types and personal preferences.

The most effective seizure detection systems are multimodal. Such systems should also be comfortable and low-power. The body signals and modalities on which a system is based should take account of the user's seizure types and personal preferences.

Nocturnal home monitoring of epileptic children is often not feasible due to the cumbersome manner of seizure monitoring with the standard method of video/EEG-monitoring. We propose a method for hypermotor seizure detection based on accelerometers attached to the extremities. From the acceleration signals, multiple temporal, frequency, and wavelet-based features are extracted. After determining the features with the highest discriminative power, we classify movement events in epileptic and nonepileptic movements. This classification is only based on a nonparametric estimate of the probability density function of normal movements. Such approach allows us to build patient-specific models to classify movement data without the need for seizure data that are rarely available. If, in the test phase, the probability of a data point (event) is lower than a threshold, this event is considered to be an epileptic seizure; otherwise, it is considered as a normal nocturnal movement event. The mean performance over seven patients gives a sensitivity of 95.24% and a positive predictive value of 60.04%. However, there is a noticeable interpatient difference.

In this study we introduce a method for detecting myoclonic jerks during the night with video. Using video instead of the traditional method of using EEG-electrodes, permits patients to sleep without any attached sensors. This improves the comfort during sleep and it makes long term home monitoring possible. The algorithm for the detection of the seizures is based on spatio-temporal interest points (STIPs), proposed by Ivan Laptev, which is the state-of-the-art in action recognition.We applied this algorithm on a group of patients suffering from myoclonic jerks. With an optimal parameter setting this resulted in a sensitivity of over 75% and a PPV of over 85%, on the patients' combined data.

The monitoring of epileptic seizures is mainly done by means of video/EEG-monitoring. Although this method is considered as the golden standard, it is not comfortable for the patient as the EEG-electrodes have to be attached to the scalp which hampers the patient's movement. This makes long term home monitoring not feasible. A detection system with accelerometers attached to the wrists and ankles can solve this problem. Nocturnal frontal lobe seizures often include bicycle pedaling movements or uncontrolled movements with the arms which are clearly visible in the accelerometer signals. Data from three patients suffering from nocturnal frontal lobe seizures is used in this paper for the development of an automatic detection algorithm for this type of seizure. First movement epochs are detected as a preprocessing step by calculating the standard deviation of a sliding window. Afterwards a moving average filter is applied to the data and thresholds are set to the signals of the arms and legs to detect the seizures. This resulted in an algorithm with a sensitivity of 91.67% and a specificity of 83.92%.

The aim of our work is to investigate whether the optical flow algorithm applied to video recordings can be used to detect movement during sleep in pediatric patients with epilepsy. The optical flow algorithm allocates intensities to pixels proportional to their involvement in movement of an object. The average of a percentage of the highest movement vectors was plotted as a function of time (R(t)). The used dataset contains video data acquired at the University Hospital of Leuven consisting of normal sleep movement and seizure movement. We investigated R(t), to make a distinction between movement and non-movement. We used the acquisition parameters (320 x 240 at 12.5 fps), derived from a previous study (Cuppens et al., Proceedings of the 4th European congress of the international federation for medical and biological engineering (MBEC 2008), ECIFBME 2008, Antwerp, Belgium, IFMBE Proceedings, vol 22, pp 784-789, 2008). Two experiments were concluded, one with global thresholds of R(t) in all datasets and one with a variable threshold in each dataset. The latter is obtained by inspecting a non-movement epoch and calculating the mean and standard deviations of R(t) over time. The variable threshold on R(t) was then obtained for each dataset by adding to the mean a fixed multiple of the standard deviation. Optimal thresholds were derived based on a three-fold cross-validation. The best result was achieved when using a variable threshold, which resulted in a sensitivity of one in all the test sets and a PPV of 1, 0.821, and 1, respectively, for the three test sets.

No abstract

A person-dependent epileptic seizure detection method has been designed that requires little human interaction. In contrast to traditional machine learning approaches, the imbalance of the dataset does not cause substantial difficulties.