Principles of time–frequency feature extraction for change detection in non-stationary signals: Applications to newborn EEG abnormality detection

Boashash, Boualem; Azemi, Ghasem; Khan, Nabeel Ali

doi:10.1016/j.patcog.2014.08.016

Cited by 104 publications

(60 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If the entropy of a TFD is to be calculated without using its singular values, the Time-Frequency Rényi entropy (TFRE) in been replaced with the corresponding twodimensional TF characteristics according to [30]. ▶ Table 3 includes all extended time-domain features.…”

Section: Entropy-based Tfd-derived Featuresmentioning

confidence: 99%

Untitled

2018

Methods Inf Med

View full text Add to dashboard Cite

SummaryObjective: This works investigates the timefrequency content of impedance cardiography signals during a propofol-remifentanil anesthesia. Materials and Methods:In the last years, impedance cardiography (ICG) is a technique which has gained much attention. However, ICG signals need further investigation. TimeFrequency Distributions (TFDs) with 5 different kernels are used in order to analyze impedance cardiography signals (ICG) before the start of the anesthesia and after the loss of consciousness. In total, ICG signals from one hundred and thirty-one consecutive patients undergoing major surgery under general anesthesia were analyzed. Several features were extracted from the calculated TFDs in order to characterize the time-frequency content of the ICG signals. Differences between those features before and after the loss of consciousness were studied. Results: The Extended Modified Beta Distribution (EMBD) was the kernel for which most features shows statistically significant changes between before and after the loss of consciousness. Among all analyzed features, those based on entropy showed a sensibility, specificity and area under the curve of the receiver operating characteristic above 60%. Conclusion:The anesthetic state of the patient is reflected on linear and non-linear features extracted from the TFDs of the ICG signals. Especially, the EMBD is a suitable kernel for the analysis of ICG signals and offers a great range of features which change according to the patient's anesthesia state in a statistically significant way.

show abstract

Section: Entropy-based Tfd-derived Featuresmentioning

confidence: 99%

Untitled

2018

Methods Inf Med

View full text Add to dashboard Cite

show abstract

“…Most of the existing abnormality detection methodologies require visual analysis by a neurophysiologist. Detection of abnormality in EEG signal is a non-stationary signal classification problem that involves extraction of features from time-domain, frequency domain or joint t-f domain representations of signal [1][2][3][4][5][6]. Recent studies have indicated that EEG signals have non-stationary characteristics, so time-frequency methods are preferred tools for their analysis [2].…”

Section: Introductionmentioning

confidence: 99%

“…Quadratic Time-Frequency Distributions (TFD) are frequently employed for analysis and extraction of features from EEG signals because of their high resolution [4]. The Wigner-Ville Distribution (WVD) is a core distribution of this class.…”

Section: Introductionmentioning

confidence: 99%

Classification of EEG Signals Using Adaptive Time-Frequency Distributions

Khan¹,

Ali²

2016

Metrology and Measurement Systems

View full text Add to dashboard Cite

Time-Frequency (t-f) distributions are frequently employed for analysis of new-born EEG signals because of their non-stationary characteristics. Most of the existing time-frequency distributions fail to concentrate energy for a multicomponent signal having multiple directions of energy distribution in the t-f domain. In order to analyse such signals, we propose an Adaptive Directional Time-Frequency Distribution (ADTFD). The ADTFD outperforms other adaptive kernel and fixed kernel TFDs in terms of its ability to achieve high resolution for EEG seizure signals. It is also shown that the ADTFD can be used to define new time-frequency features that can lead to better classification of EEG signals, e.g. the use of the ADTFD leads to 97.5% total accuracy, which is by 2% more than the results achieved by the other methods.

show abstract

“…Much research focuses on specific patterns: for example, neonatal seizure detection and localisation [29,[35][36][37], and detection of the sleep-wake cycle [38]. Classification of a broader variety of EEG background patterns will increase the system complexity and the need for specific EEG data, especially in long-term multichannel EEG recordings.…”

Section: Significance and Motivationmentioning

confidence: 99%

“…Most research has focused on specific patterns: for example, seizure detection and localisation [29,[35][36][37], burst-suppression classification [24,[152][153][154] and detection of the sleep-wake cycle [38]. Classification of a broader variety of EEG background patterns will increase the system complexity.…”

Section: Visual Inspection Of the Eeg Signal Is Routine But Visual Inmentioning

confidence: 99%

Design and optimisation of time-frequency analysis for multichannel neonatal EEG background features in term neonates with hypoxic ischaemic encephalopathy: characterisation, classification and neurodevelopmental outcome prediction

Awal¹

View full text Add to dashboard Cite

Neonatal Hypoxic Ischaemic Encephalopathy (HIE) is a major cause of morbidity and mortality in newborns throughout the world with an incidence of 1-8/1000 newborns in developed countries and up to 26/1000 newborns in developing countries. It causes one million neonatal deaths globally per year. Electroencephalography (EEG) is a useful method for assessment of brain activity in newborns with HIE. EEG performs well in the early diagnosis and classification of HIE severity and in predicting neurodevelopmental outcome. A normal EEG is highly predictive of a normal outcome and various abnormal EEG features are associated with neurological abnormalities or death.A systematic review and meta-analysis of the literature was undertaken to determine which specific background features of the EEG best predict outcome (Aim 1). Automatic detection and classification

show abstract

Principles of time–frequency feature extraction for change detection in non-stationary signals: Applications to newborn EEG abnormality detection

Cited by 104 publications

References 37 publications

Untitled

Untitled

Classification of EEG Signals Using Adaptive Time-Frequency Distributions

Design and optimisation of time-frequency analysis for multichannel neonatal EEG background features in term neonates with hypoxic ischaemic encephalopathy: characterisation, classification and neurodevelopmental outcome prediction

Contact Info

Product

Resources

About