g i-, d y s i s and Optimized Design of Single Feed Circularly Polarized , Microstrip Antennas .. Absfract-Analysis and optimized gesigw are presented of three types k a 4
The detection of nonconvulsive seizures (NCSz) is a challenge because of the lack of physical symptoms, which may delay the diagnosis of the disease. Many researchers have reported automatic detection of seizures. However, few investigators have concentrated on detection of NCSz. This article proposes a method for reliable detection of NCSz. The electroencephalography (EEG) signal is usually contaminated by various nonstationary noises. Signal denoising is an important preprocessing step in the analysis of such signals. In this study, a new wavelet-based denoising approach using cubical thresholding has been proposed to reduce noise from the EEG signal prior to analysis. Three statistical features were extracted from wavelet frequency bands, encompassing the frequency range of 0 to 8, 8 to 16, 16 to 32, and 0 to 32 Hz. Extracted features were used to train linear classifier to discriminate between normal and seizure EEGs. The performance of the method was tested on a database of nine patients with 24 seizures in 80 hours of EEG recording. All the seizures were successfully detected, and false positive rate was found to be 0.7 per hour.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.