Mallat’s Scattering Transform Based Anomaly Sensing for Detection of Seizures in Scalp EEG

Ahmad, Maham; Kamboh, Awais Mehmood; Saleem, Sajid; Khan, Amir

doi:10.1109/access.2017.2736014

Cited by 29 publications

(11 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A multivariate feature extraction approach was adopted to extract textual features, univariate, bivariate, and multivariate features extracted using channel selection; these features were mapped to the 2D image, and the GLCM matrix was applied to extract homogeneity features [29]. Mallet's scattering transform was applied to extract Shannon entropy, Renyi entropy, permutation, and spectral entropies [53]. Eight absolute spectral power features and relative spectral power features, spectral power ratio features of 44 features were extracted by employing the sparse feature selection method, in particular, sparse Bayesian multinomial logistic regression (SBMLR), which increases classification accuracy [60].…”

Section: Other Feature Extraction Methodsmentioning

confidence: 99%

“…Energy components were extracted over the delta, theta, alpha, beta, gamma1, and gamma2 frequency bands via calculating PSD by incorporating Fast Fourier Transform (FFT), and additionally, DWT was applied to extract seven-level decomposition coefficients [69]. The scattering transform and DWT were adapted to perform feature extraction and extracted 45 features related to spectra, entropies, Hurst exponent, line length, power spectra, and fractal dimensions [53].…”

Section: Continuous Wavelet Transform (Cwt)mentioning

confidence: 99%

“…The extracted features were fed into the SVM classifier to perform the classification [52]. An unsupervised method of predicting seizures was used to perform a classification with the mallet scattering transform to analyze an EEG signal, which attained a specificity of 98% and a sensitivity of 78% [53]. A multivariate method of empirical wavelet transform (EWT) was performed to extract the characteristics, and different classifiers were used for classification, which achieved a sensitivity of 97.91%, specificity of 99.41%, and precision of 99.41% [54].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Automated Epileptic Seizure Detection in Pediatric Subjects of CHB-MIT EEG Database—A Survey

Prasanna

Subathra

Mohammed

et al. 2021

JPM

View full text Add to dashboard Cite

Epilepsy is a neurological disorder of the brain that causes frequent occurrence of seizures. Electroencephalography (EEG) is a tool that assists neurologists in detecting epileptic seizures caused by an unexpected flow of electrical activities in the brain. Automated detection of an epileptic seizure is a crucial task in diagnosing epilepsy which overcomes the drawback of a visual diagnosis. The dataset analyzed in this article, collected from Children’s Hospital Boston (CHB) and the Massachusetts Institute of Technology (MIT), contains long-term EEG records from 24 pediatric patients. This review paper focuses on various patient-dependent and patient-independent personalized medicine approaches involved in the computer-aided diagnosis of epileptic seizures in pediatric subjects by analyzing EEG signals, thus summarizing the existing body of knowledge and opening up an enormous research area for biomedical engineers. This review paper focuses on the features of four domains, such as time, frequency, time-frequency, and nonlinear features, extracted from the EEG records, which were fed into several classifiers to classify between seizure and non-seizure EEG signals. Performance metrics such as classification accuracy, sensitivity, and specificity were examined, and challenges in automatic seizure detection using the CHB-MIT database were addressed.

show abstract

Section: Other Feature Extraction Methodsmentioning

confidence: 99%

Section: Continuous Wavelet Transform (Cwt)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automated Epileptic Seizure Detection in Pediatric Subjects of CHB-MIT EEG Database—A Survey

Prasanna

Subathra

Mohammed

et al. 2021

JPM

View full text Add to dashboard Cite

show abstract

“…Another method that researchers have used is the group invariant scattering transform. Feature extraction using this method has been performed by Ahmad et al [20] and the algorithm was evaluated on the CHB-MIT dataset. An accuracy of 91% is reported using a semi-supervised anomaly detection approach.…”

Section: Related Workmentioning

confidence: 99%

Automatic Seizure Detection Using Multi-Resolution Dynamic Mode Decomposition

et al. 2019

Self Cite

View full text Add to dashboard Cite

Epilepsy is one of the most prevalent neurological issues faced by a large population around the globe. Epilepsy is marked by intermittent seizures, the detection of which can be a challenging problem. Therefore, reliably detecting the onset of seizures has evoked the interest of researchers over the last few years. Major leaps in the domain of machine learning, signal processing methods, and computational capabilities have made it a tractable task. In this paper, we apply multi-resolution dynamic mode decomposition (MRDMD), which is a data-driven dimensionality reduction technique, on the problem of epileptic seizure detection. This method can effectively separate a complex non-linear system into a collection of timescale components at different resolutions. We have applied this algorithm on two different scalp EEG datasets, i.e., CHB-MIT and KU Leaven datasets. We have applied necessary post-processing steps to reduce the false alarm rate and boost the sensitivity and specificity. A detailed analysis of the results has been presented for the proposed method applied to both the datasets. The algorithm achieves a sensitivity of 0.937 and 0.96, a specificity of 0.99 and 0.99, a false alarm rate of 0.587 and 0.413 per hour, and a latency of 3.12 and 2.75 s for CHB-MIT and KU Leuven, respectively. The results indicate that the multi-resolution analysis yields a significant improvement in sensitivity compared with the basic dynamic mode decomposition.

show abstract

“…The EEG signals are amplified during recording, after which different signal processing techniques from both time and frequency domain analysis could be used for extracting the required features, such as wavelet transform [16], empirical mode decomposition [17], [18], and scattering transform [19]- [23], and dynamic mode decomposition [24], [25], etc. For analysis of these signals, different machine learning algorithms like support-vector machine (SVM), principal component analysis (PCA), lineardiscriminant analysis (LDA), neural networks, decision-tree classifiers, and sometimes a combination of these techniques are employed [26]- [31].…”

Section: Introductionmentioning

confidence: 99%

Epileptic Seizure Detection With a Reduced Montage: A Way Forward for Ambulatory EEG Devices

et al. 2020

Self Cite

View full text Add to dashboard Cite

Electroencephalogram (EEG) is one of the fundamental tools for analyzing the behavior of brain and particularly helpful for treatment of epilepsy and detection of associated seizures. For longterm recording of EEG signals, current research is heading towards simple, unobtrusive and ambulatory devices with a small number of channels. The primary contribution of this paper is to assess the performance difference between the seizure detection results using features from all channels versus only the channels in/around the temporal region. For this purpose, we develop a supervised seizure detection algorithm that uses time domain features extracted sequentially for every 1-second epoch. By using this algorithm, we obtained sensitivity values of 0.95 and 0.92, specificity values of 0.99 and 0.99 and false positive per hour values as 0.16 and 0.21 for all 23 channels and 10 temporal region channels, respectively. These results show that restricting the EEG analysis to temporal region results only in a graceful and gradual degradation of classifier performance. We conclude that EEG ambulatory devices with a montage local to the temporal region could demonstrate satisfactory performance. This presents a promising way forward for the use of ambulatory devices with compact wearable design.

show abstract

Mallat’s Scattering Transform Based Anomaly Sensing for Detection of Seizures in Scalp EEG

Cited by 29 publications

References 34 publications

Automated Epileptic Seizure Detection in Pediatric Subjects of CHB-MIT EEG Database—A Survey

Automated Epileptic Seizure Detection in Pediatric Subjects of CHB-MIT EEG Database—A Survey

Automatic Seizure Detection Using Multi-Resolution Dynamic Mode Decomposition

Epileptic Seizure Detection With a Reduced Montage: A Way Forward for Ambulatory EEG Devices

Contact Info

Product

Resources

About