EEG seizure detection by integrating slantlet transform with sparse coding

Abstract

EEG signals, recording both normal and abnormal activities of neurons in the brain, are widely used in epilepsy detection. In this paper, an EEG signal classification method based on Slantlet Transform and sparse coding is proposed to greatly reduce number of false alarms and improve speed of detection. With Slantlet Transform, salient information of EEG signals is mapped into a sparse space. In order to ensure good detection rates from the EEG signals, a Sparse Representation Classifier is adopted. Experimental results demonstrate that the in terms of detection rate, the proposed method, which integrates Slantlet Transform with sparse coding, outperforms the current state-of-the-art EEG seizure detection methods.

Neural network based seizure detection system using statistical package analysis

“…The similar finding and this similar finding for the related study discussed as, proposed the patient-specific real-time automatic epileptic seizure detection system using long term scalp and short-term scalp data and achieved the accuracy of 96% sensitivity, 0.1 per hour median false detection rate using ANN algorithm. Hou et al [29] proposed the epilepsy detection system for locating the epileptic foci in the brain using random forest classification model using locally linear embedding algorithm and the optimization improved 0.95 of effective result when compared with SVM, decision tree, KNN, and random forest.…”

StationPlot: A New Non-stationarity Quantification Tool for Detection of Epileptic Seizures

“…Further, joint time-frequency techniques like wavelet transform & machine learning classifiers [6,7], intra-cranial spatiotemporal correlation structures [8], time-frequency image descriptors [9] & multi-level wavelet decomposition [10] have shown potency in epileptic seizure diagnosis. Non-linear methods like permutation entropy & support vector machines (SVM) [11], multi-scale sample entropy [12], rational discrete short-time Fourier transform [13], sparse coding using Slantlet transform [14] for multi-resolution quantification & sparse representation of seizure free (interictal) & seizure (ictal) EEG signals is studied in [15]. Application of local pattern transformation *This paper is accepted for presentation at IEEE Global Conference on Signal and Information Processing (IEEE GlobalSIP), California, USA, 2018 features [16], ARIMA-GARCH models [17] & L l regularized EEG attributes [18] have shown adequacy as potential SA biomarkers.…”