2015
DOI: 10.1016/j.yebeh.2015.06.002
|View full text |Cite
|
Sign up to set email alerts
|

Band-sensitive seizure onset detection via CSP-enhanced EEG features

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

1
20
0

Year Published

2016
2016
2023
2023

Publication Types

Select...
4
2
1

Relationship

1
6

Authors

Journals

citations
Cited by 32 publications
(21 citation statements)
references
References 40 publications
1
20
0
Order By: Relevance
“…Each selected time period was analyzed using time segments of 200 ms. Common spatial pattern (CSP) is widely used and highly successful in the binary case [33], which was applied in this study to perform feature extraction. Given two distributions in a high-dimensional space, the CSP algorithm can calculate spatial filters that maximize the variance between two classes [34]. For the analysis, let V denote the raw data of a single trial, an N×T matrix with N the number of channels (28 channels in this study) and T the number of samples in time.…”
Section: Methodsmentioning
confidence: 99%
“…Each selected time period was analyzed using time segments of 200 ms. Common spatial pattern (CSP) is widely used and highly successful in the binary case [33], which was applied in this study to perform feature extraction. Given two distributions in a high-dimensional space, the CSP algorithm can calculate spatial filters that maximize the variance between two classes [34]. For the analysis, let V denote the raw data of a single trial, an N×T matrix with N the number of channels (28 channels in this study) and T the number of samples in time.…”
Section: Methodsmentioning
confidence: 99%
“…We also report latency (time delay) between the annotated seizure onset and the detection by the method in Table 3. We compare classification accuracy and latency with the state-ofthe-art methods [13,26,23], which also report classification performance and latency for the Children's Hospital Boston database. We emphasise again that these state-of-the-art methods are significantly more computationally expensive than the proposed method.…”
Section: Resultsmentioning
confidence: 99%
“…Most methods use classification techniques from the supervised machine learning literature, such as support vector machines [10][11][12] and discriminant analysis [13], and differ mainly in terms of their feature extraction methods and the features classification approaches. Many methods use time-frequency descriptors, either explicitly (e.g., short-term Fourier or wavelet representations) [14][15][16][17]11,18,19], empirical mode decomposition [20][21][22], or implicitly by learning neural networks [23][24][25] or by using component analysis or common spatial patterns (see for example [26][27][28]). Some also use statistical descriptors such as signal entropy [17,11,[29][30][31] or fractal dimension [32,33].…”
Section: Introductionmentioning
confidence: 99%
See 2 more Smart Citations