Comparison of STFT and wavelet transform methods in determining epileptic seizure activity in EEG signals for real-time application

Kim

et al. 2016

Human Brain Mapping

Few studies have directly compared the neural correlates of spatial attention (i.e., attention to a particular location) and nonspatial attention (i.e., attention to a feature in the visual scene) using well‐controlled tasks. Here, we investigated the neural correlates of spatial and nonspatial attention in humans using intracranial electroencephalography. The topography and number of electrodes showing significant event‐related desynchronization (ERD) or event‐related synchronization (ERS) in different frequency bands were studied in 13 epileptic patients. Performance was not significantly different between the two conditions. In both conditions, ERD in the low‐frequency bands and ERS in the high‐frequency bands were present bilaterally in the parietal cortex (prominently on the right hemisphere) and frontal regions. In addition to these common changes, spatial attention involved right‐lateralized activity that was maximal in the right superior parietal lobule (SPL), whereas nonspatial attention involved wider brain networks including the bilateral parietal, frontal, and temporal regions, but still had maximal activity in the right parietal lobe. Within the parietal lobe, spatial attention involved ERD or ERS in the right SPL, whereas nonspatial attention involved ERD or ERS in the right inferior parietal lobule. These findings reveal that common as well as different brain networks are engaged in spatial and nonspatial attention. Hum Brain Mapp 37:3041–3054, 2016. © 2016 The Authors Human Brain Mapping Published by Wiley Periodicals, Inc.

Section: Methodsmentioning

confidence: 99%

Neural correlates of spatial and nonspatial attention determined using intracranial electroencephalographic signals in humans

Kim

et al. 2016

Human Brain Mapping

J NeuroEngineering Rehabil

“…The non-parametric TFRs such as band-pass filtering, short time Fourier transform (STFT) and continuous wavelet transform (CWT) were successfully applied in time-frequency analysis of EEG [20,21]. However, all the traditional methods have pros and cons in temporal and spectral resolutions.…”

Section: Introductionmentioning

confidence: 99%

Time-frequency analysis of band-limited EEG with BMFLC and Kalman filter for BCI applications

Wang

Veluvolu

2013

BackgroundTime-Frequency analysis of electroencephalogram (EEG) during different mental tasks received significant attention. As EEG is non-stationary, time-frequency analysis is essential to analyze brain states during different mental tasks. Further, the time-frequency information of EEG signal can be used as a feature for classification in brain-computer interface (BCI) applications.MethodsTo accurately model the EEG, band-limited multiple Fourier linear combiner (BMFLC), a linear combination of truncated multiple Fourier series models is employed. A state-space model for BMFLC in combination with Kalman filter/smoother is developed to obtain accurate adaptive estimation. By virtue of construction, BMFLC with Kalman filter/smoother provides accurate time-frequency decomposition of the bandlimited signal.ResultsThe proposed method is computationally fast and is suitable for real-time BCI applications. To evaluate the proposed algorithm, a comparison with short-time Fourier transform (STFT) and continuous wavelet transform (CWT) for both synthesized and real EEG data is performed in this paper. The proposed method is applied to BCI Competition data IV for ERD detection in comparison with existing methods.ConclusionsResults show that the proposed algorithm can provide optimal time-frequency resolution as compared to STFT and CWT. For ERD detection, BMFLC-KF outperforms STFT and BMFLC-KS in real-time applicability with low computational requirement.

J Med Imaging Hlth Inform

“…22 fixed resolution limitation of STFT, the wavelet theory was invented. [22][23][24] The Wavelet Transform (WT) provides a flexible timefrequency window: it narrows when it analyzes high frequencies and widens when it detects low frequencies. 22 It must be emphasized that the Fourier and Wavelet transforms are linear time-frequency representations since they decompose a given signal on elementary components: the atoms.…”

Section: Introductionmentioning

confidence: 99%

2D DNA Representations Generated Using a New Coding and the Time-Frequency Analysis

Messaoudi¹,

Oueslati²,

Lachiri³

2015

The state of the art shows that Fourier transform based methods lack robustness in terms of the DNA characterization and detection. In fact, the Fourier analysis is limited by the shape and the fixed size of the analysis window; which can affect the representation resolution. In this paper, we review the power of the time-frequency analysis in characterizing the biological sequences. In particular, we study the performance of the wavelet analysis in distinguishing specific structures within the C. elegans genome. The method offers the possibility to visualize the genomic data within the form of explicit images in which DNA elements possess particular behavior. And thus, we can clearly identify homologous sequences through their common behavior. A comparison with the smoothed Fourier analysis and the Choi-Williams distribution is then established. Even the Smoothed Fourier technique is limited as it is a fixed resolution method; thereby the multi-resolution analysis is proposed to improve the DNA analysis. On the other hand, the Choi-Williams distribution increases the time-frequency resolution. Nevertheless, it exhibits the presence of cross-terms which can be considered as signal components. Most of all, one must digitize the DNA characters to be able to use the proposed tools. In line with this, we use the Frequency Chaos Game Signal (FCGS) as coding technique. The FCGS is built in such way that we can follow the frequency evolution of nucleotides' occurrence along the genome. Through this work, we demonstrate that color-scalogram is more effective than the spectrogram and the Choi-Williams representations; since it enables the characterization of different DNA types by specific texture. Besides the visual inspection of the representations, the Shannon entropy measure proves the effectiveness of the wavelet method.