Epileptic seizure detection on a compressed EEG signal using energy measurement

Wijayanto, Inung; Humairani, Annisa; Hadiyoso, Sugondo; Rizal, Achmad; Prasanna, Dasari Lakshmi; Tripathi, Suman Lata

doi:10.1016/j.bspc.2023.104872

Cited by 21 publications

(3 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The time evolution of fractal phenomena, as well as the roughness of the related time series, can be identified using the Hurst exponent [1,42,43]. Various research topics have been investigated with the use of Hurst exponent, such as hydrology [40,41], astrophysics and applications [44,45], processes of capital markets [46][47][48][49], noisy observations of traces in traffic [50][51][52], seizures prior to epilepsy [53][54][55], climatic dynamics [56], and precursory time series before impending earthquakes [1,7,8,57].…”

Section: Hurst Exponentmentioning

confidence: 99%

Fractal Patterns in Groundwater Radon Disturbances Prior to the Great 7.9 Mw Wenchuan Earthquake, China

Alam,

Nikolopoulos,

Wang

2023

Geosciences

View full text Add to dashboard Cite

This study reports a fractal analysis of one-year radon in groundwater disturbances from five stations in China amidst the catastrophic Wenchuan (Mw = 7.9) earthquake of 12 May 2008 (day 133). Five techniques are used (DFA, fractal dimensions with Higuchi, Katz, Sevcik methods, power-law analysis) in segmented portions glided throughout each signal. Noteworthy fractal areas are outlined in the KDS, GS, MSS data, whilst the portions were non-significant for PZHS and SPS. Up to day 133, critical epoch DFA-exponents are 1.5≤α<2.0, with several above 1.8. The fractal dimensions exhibit Katz’s D around 1.0–1.2, Higuchi’s D between 1.5 and 2.0, and Sevcik’s D between 1.0 and 1.5. Several power-law exponents are above 1.7, and numerous are above 2.0. All fractal results of the KDS-GS-MSS are further analysed using a novel computerised methodology that locates the exact out-of-threshold fractal areas and combines the outcomes of different methods per five, four, three, and two (maximum 13 combinations) versus nineteen Mw≥ 5.5 earthquakes of the greater area. Most coincidences using different techniques are before the great Wenchuan earthquake and after the earthquake. This is not only with one method but with 13 different methods. Other interpretations are also discussed.

show abstract

Section: Hurst Exponentmentioning

confidence: 99%

Fractal Patterns in Groundwater Radon Disturbances Prior to the Great 7.9 Mw Wenchuan Earthquake, China

Alam,

Nikolopoulos,

Wang

2023

Geosciences

View full text Add to dashboard Cite

show abstract

“…Compression sensing (CS) has been widely used to compress physiological signals in the past decades [16,17]. It projects signals into a sparse domain, using random sampling at a much lower rate than the Nyquist sampling rate, and then reconstructs the signal through a non-linear reconstruction algorithm [18]. Although CS has proven its effective, it has limitations when applied to EEG signal compression.…”

Section: Introductionmentioning

confidence: 99%

A feature enhanced EEG compression model using asymmetric encoding–decoding network ^*

Wang,

Zhang,

2024

J. Neural Eng.

View full text Add to dashboard Cite

Objective. Recently, the demand for wearable devices using electroencephalography (EEG) has increased rapidly in many fields. Due to its volume and computation constraints, wearable devices usually compress and transmit EEG to external devices for analysis. However, current EEG compression algorithms are not tailor-made for wearable devices with limited computing and storage. Firstly, the huge amount of parameters makes it difficult to apply in wearable devices; secondly, it’s tricky to learn EEG signals’ distribution law due to the low signal-to-noise ratio, which leads to excessive reconstruction error and suboptimal compression performance. Approach. Here, a feature enhanced asymmetric encoding-decoding network is proposed. EEG is encoded with a lightweight model, and subsequently decoded with a multi-level feature fusion network by extracting the encoded features deeply and reconstructing the signal through a two-branch structure. Main results. On public EEG datasets, motor imagery and event-related potentials, experimental results show that the proposed method has achieved the state of the art compression performance. In addition, the neural representation analysis and the classification performance of the reconstructed EEG signals also show that our method tends to retain more task-related information as the compression ratio increases and retains reliable discriminative information after EEG compression. Significance. This paper tailors an asymmetric EEG compression method for wearable devices that achieves state-of-the-art compression performance in a lightweight manner, paving the way for the application of EEG-based wearable devices.

show abstract

“…There are many application examples of these methods in different areas. For example, in medicine, FR improves the accuracy of epilepsy diagnosis through analyzing electroencephalography signals, avoiding invasive techniques [9]. FS has been also used for breast cancer detection [10] analyzing microarray data.…”

Section: Introductionmentioning

confidence: 99%

An Extensive Performance Comparison between Feature Reduction and Feature Selection Preprocessing Algorithms on Imbalanced Wide Data

Ramos-Pérez,

Barbero-Aparicio,

Canepa-Oneto

et al. 2024

Information

View full text Add to dashboard Cite

The most common preprocessing techniques used to deal with datasets having high dimensionality and a low number of instances—or wide data—are feature reduction (FR), feature selection (FS), and resampling. This study explores the use of FR and resampling techniques, expanding the limited comparisons between FR and filter FS methods in the existing literature, especially in the context of wide data. We compare the optimal outcomes from a previous comprehensive study of FS against new experiments conducted using FR methods. Two specific challenges associated with the use of FR are outlined in detail: finding FR methods that are compatible with wide data and the need for a reduction estimator of nonlinear approaches to process out-of-sample data. The experimental study compares 17 techniques, including supervised, unsupervised, linear, and nonlinear approaches, using 7 resampling strategies and 5 classifiers. The results demonstrate which configurations are optimal, according to their performance and computation time. Moreover, the best configuration—namely, k Nearest Neighbor (KNN) + the Maximal Margin Criterion (MMC) feature reducer with no resampling—is shown to outperform state-of-the-art algorithms.

show abstract

Epileptic seizure detection on a compressed EEG signal using energy measurement

Cited by 21 publications

References 24 publications

Fractal Patterns in Groundwater Radon Disturbances Prior to the Great 7.9 Mw Wenchuan Earthquake, China

Fractal Patterns in Groundwater Radon Disturbances Prior to the Great 7.9 Mw Wenchuan Earthquake, China

A feature enhanced EEG compression model using asymmetric encoding–decoding network ^*

An Extensive Performance Comparison between Feature Reduction and Feature Selection Preprocessing Algorithms on Imbalanced Wide Data

Contact Info

Product

Resources

About

Epileptic seizure detection on a compressed EEG signal using energy measurement

Cited by 21 publications

References 24 publications

Fractal Patterns in Groundwater Radon Disturbances Prior to the Great 7.9 Mw Wenchuan Earthquake, China

Fractal Patterns in Groundwater Radon Disturbances Prior to the Great 7.9 Mw Wenchuan Earthquake, China

A feature enhanced EEG compression model using asymmetric encoding–decoding network *

An Extensive Performance Comparison between Feature Reduction and Feature Selection Preprocessing Algorithms on Imbalanced Wide Data

Contact Info

Product

Resources

About

A feature enhanced EEG compression model using asymmetric encoding–decoding network ^*