No abstract

No abstract

No abstract

Epilepsy is the second most common brain disorder after migraine. Automatic detection of epileptic seizures can considerably improve the patients' quality of life. Current Electroencephalogram (EEG)-based seizure detection systems encounter many challenges in real-life situations. The EEGs are non-stationary signals and seizure patterns vary across patients and recording sessions. Moreover, EEG data are prone to numerous noise types that negatively affect the detection accuracy of epileptic seizures. To address these challenges, we introduce the use of a deep learning-based approach that automatically learns the discriminative EEG features of epileptic seizures. Specifically, to reveal the correlation between successive data samples, the time-series EEG data are first segmented into a sequence of nonoverlapping epochs. Second, Long Short-Term Memory (LSTM) network is used to learn the high-level representations of the normal and the seizure EEG patterns. Third, these representations are fed into Softmax function for training and classification. The results on a well-known benchmark clinical dataset demonstrate the superiority of the proposed approach over the existing stateof-the-art methods. Furthermore, our approach is shown to be robust in noisy and real-life conditions. Compared to current methods that are quite sensitive to noise, the proposed method maintains its high detection performance in the presence of common EEG artifacts (muscle activities and eye-blinking) as well as white noise.

No abstract

No abstract

Brain is the most important part in the human body controlling muscles and nerves; Electroencephalogram (EEG) signals record brain electric activities. EEG signals capture important information pertinent to different physiological brain states. In this paper, we propose an efficient framework for evaluating the power-accuracy trade-off for EEG-based compressive sensing and classification techniques in the context of epileptic seizure detection in wireless tele-monitoring. The framework incorporates compressive sensing-based energyefficient compression, and noisy wireless communication channel to study the effect on the application accuracy. Discrete cosine transform (DCT) and compressive sensing are used for EEG signals acquisition and compression. To obtain low-complexity energy-efficient, the best data accuracy with higher compression ratio is sought. A reconstructed algorithm derived from DCT of daubechie's wavelet 6 is used to decompose the EEG signal at different levels. DCT is combined with the best basis function neural networks for EEG signals classification. Extensive experimental work is conducted, utilizing four classification models. The obtained results show an improvement in classification accuracies and an optimal classification rate of about 95% is achieved when using NN classifier at 85% of CR in the case of no SNR value. The satisfying results demonstrate the effect of efficient compression on maximizing the sensor lifetime without affecting the application's accuracy.

No abstract