Classification of EEG Signals Using Adaptive Time-Frequency Distributions

Khan, Nabeel Ali; Ali, Sadiq

doi:10.1515/mms-2016-0021

Cited by 22 publications

(24 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Classification of ECoG signals plays an important role in many neurological studies, for example, for staging a neurologic disease or for brain-machine interface applications [29,30]. In this study, PCA was applied to ECoG, for highlighting how rats' brain codes different stimulus features [31].…”

Section: Discussionmentioning

confidence: 99%

Biomimetic Sonar for Electrical Activation of the Auditory Pathway

et al. 2017

View full text Add to dashboard Cite

Relying on the mechanism of bat's echolocation system, a bioinspired electronic device has been developed to investigate the cortical activity of mammals in response to auditory sensorial stimuli. By means of implanted electrodes, acoustical information about the external environment generated by a biomimetic system and converted in electrical signals was delivered to anatomically selected structures of the auditory pathway. Electrocorticographic recordings showed that cerebral activity response is highly dependent on the information carried out by ultrasounds and is frequency-locked with the signal repetition rate. Frequency analysis reveals that delta and beta rhythm content increases, suggesting that sensorial information is successfully transferred and integrated. In addition, principal component analysis highlights how all the stimuli generate patterns of neural activity which can be clearly classified. The results show that brain response is modulated by echo signal features suggesting that spatial information sent by biomimetic sonar is efficiently interpreted and encoded by the auditory system. Consequently, these results give new perspective in artificial environmental perception, which could be used for developing new techniques useful in treating pathological conditions or influencing our perception of the surroundings. BackgroundUltrasound (US) plays an important role in the environmental perception of many species of mammals (such as rats and bats) for communication and survival; indeed, they use incoming US to gather information about incumbent danger, food availability, and navigational ranging [1]. Although Rodentia and Microchiroptera do not claim a common phylogenetic origin, recent literature reports that there are certain similarities in both mammals regarding anatomical organization [2,3]. Moreover, the cortical area arrangement and cortical frequency processing are similar to other mammals [4][5][6].US frequency range of bats and rats partially overlaps and both eutherian lineages possess similar encoding process of mechanical US waves, although with different aims as spatial navigation and social communication [7,8]. In this study, a method for stimulating rats neural centres by using a bats bioinspired neuroelectronic interface is proposed. The electronic system for US signal processing is based on the natural sonar of Pteronotus parnellii and was previously investigated for modulating brain activity [9,10]. This study demonstrated ultrasound waves, which are properly decoded into bioelectrical signals, travelled along the acoustic pathway, and successfully reached cortical areas, processing sensorial information according to the environment or to cognitive demands. Similarly, exogenous stimulation may alter sensory perception by acting on the neural networks [11].Hereafter, we have investigated how different signal patterns, emulating ultrasound echoes, affected brain activity by means of implanted electrodes connected to an electronic interface. Different patterns of echo signals at 50 ...

show abstract

Section: Discussionmentioning

confidence: 99%

Biomimetic Sonar for Electrical Activation of the Auditory Pathway

et al. 2017

View full text Add to dashboard Cite

show abstract

“…Khan and Ali () proposed a time‐frequency domain based categorization of EEG signals. The authors achieved 97.5% of an average accuracy for the signals classifications.…”

Section: Related Workmentioning

confidence: 99%

Classification of focal and nonfocal EEG signals using ANFIS classifier for epilepsy detection

Deivasigamani

Senthilpari

Yong

2016

Int. J. Imaging Syst. Technol.

View full text Add to dashboard Cite

The electroencephalogram (EEG) is the frequently used signal to detect epileptic seizures in the brain. For a successful epilepsy surgery, it is very essential to localize epileptogenic area in the brain. The signals from the epileptogenic area are focal signals and signals from other area of the brain region nonfocal signals. Hence, the classification of focal and nonfocal signals is important for locating the epileptogenic area for epilepsy surgery. In this article, we present a computer aided automatic detection and classification method for focal and nonfocal EEG signal. The EEG signal is decomposed by Dual Tree Complex Wavelet Transform (DT-CWT) and the features are computed from the decomposed coefficients. These features are trained and classified using Adaptive Neuro Fuzzy Inference System (ANFIS) classifier. The proposed system achieves 98% sensitivity, 100% specificity, and 99% accuracy for EEG signal classification. The experimental results are presented to show the effectiveness of the proposed classification method to classify the focal and nonfocal EEG signals.

show abstract

“…Previous studies have shown that the ADTFD offers high resolution and can be used to estimate the instantaneous frequency of closely spaced signal components [33]. Due to its high resolution, the ADTFD has found applications in e.g., classification of EEG signals and direction of arrival estimation [34,35].…”

Section: Introductionmentioning

confidence: 99%

Performance Comparison of Time-Frequency Distributions for Estimation of Instantaneous Frequency of Heart Rate Variability Signals

2017

View full text Add to dashboard Cite

Abstract:The instantaneous frequency (IF) of a non-stationary signal is usually estimated from a time-frequency distribution (TFD). The IF of heart rate variability (HRV) is an important parameter because the power in a frequency band around the IF can be used for the interpretation and analysis of the respiratory rate but also for a more accurate analysis of heart rate (HR) signals. In this study, we compare the performance of five states of the art kernel-based time-frequency distributions (TFDs) in terms of their ability to accurately estimate the IF of HR signals. The selected TFDs include three widely used fixed kernel methods: the modified B distribution, the S-method and the spectrogram; and two adaptive kernel methods: the adaptive optimal kernel TFD and the recently developed adaptive directional TFD. The IF of the respiratory signal, which is usually easier to estimate as the respiratory signal is a mono-component with small amplitude variations with time, is used as a reference to examine the accuracy of the HRV IF estimates. Experimental results indicate that the most reliable estimates are obtained using the adaptive directional TFD in comparison to other commonly used methods such as the adaptive optimal kernel TFD and the modified B distribution.

show abstract

Classification of EEG Signals Using Adaptive Time-Frequency Distributions

Cited by 22 publications

References 16 publications

Biomimetic Sonar for Electrical Activation of the Auditory Pathway

Biomimetic Sonar for Electrical Activation of the Auditory Pathway

Classification of focal and nonfocal EEG signals using ANFIS classifier for epilepsy detection

Performance Comparison of Time-Frequency Distributions for Estimation of Instantaneous Frequency of Heart Rate Variability Signals

Contact Info

Product

Resources

About