Matches in SemOpenAlex for { <https://semopenalex.org/work/W2617361225> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W2617361225 endingPage "67" @default.
- W2617361225 startingPage "57" @default.
- W2617361225 abstract "Epilepsy forecasting has been extensively studied using high-order time series obtained from scalp-recorded electroencephalography (EEG). An accurate seizure prediction system would not only help significantly improve patients’ quality of life, but would also facilitate new therapeutic strategies to manage epilepsy. This paper thus proposes an improved Kalman Filter (KF) algorithm to mine seizure forecasts from neural activity by modeling three properties in the high-order EEG time series: noise, temporal smoothness, and tensor structure. The proposed High-Order Kalman Filter (HOKF) is an extension of the standard Kalman filter, for which higher-order modeling is limited. The efficient dynamic of HOKF system preserves the tensor structure of the observations and latent states. As such, the proposed method offers two main advantages: (i) effectiveness with HOKF results in hidden variables that capture major evolving trends suitable to predict neural activity, even in the presence of missing values; and (ii) scalability in that the wall clock time of the HOKF is linear with respect to the number of time-slices of the sequence. The HOKF algorithm is examined in terms of its effectiveness and scalability by conducting forecasting and scalability experiments with a real epilepsy EEG dataset. The results of the simulation demonstrate the superiority of the proposed method over the original Kalman Filter and other existing methods." @default.
- W2617361225 created "2017-06-05" @default.
- W2617361225 creator A5051371429 @default.
- W2617361225 creator A5062627428 @default.
- W2617361225 creator A5087619194 @default.
- W2617361225 date "2017-08-01" @default.
- W2617361225 modified "2023-10-18" @default.
- W2617361225 title "HOKF: High Order Kalman Filter for Epilepsy Forecasting Modeling" @default.
- W2617361225 cites W1747831670 @default.
- W2617361225 cites W1814521481 @default.
- W2617361225 cites W1987766637 @default.
- W2617361225 cites W2008114373 @default.
- W2617361225 cites W2020713213 @default.
- W2617361225 cites W2024165284 @default.
- W2617361225 cites W2035503723 @default.
- W2617361225 cites W2053744708 @default.
- W2617361225 cites W2055724818 @default.
- W2617361225 cites W2061370212 @default.
- W2617361225 cites W2071194462 @default.
- W2617361225 cites W2105934661 @default.
- W2617361225 cites W2119412403 @default.
- W2617361225 cites W2120853742 @default.
- W2617361225 cites W2145341337 @default.
- W2617361225 cites W2159882659 @default.
- W2617361225 cites W2284031736 @default.
- W2617361225 cites W2403175639 @default.
- W2617361225 doi "https://doi.org/10.1016/j.biosystems.2017.02.004" @default.
- W2617361225 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/28571930" @default.
- W2617361225 hasPublicationYear "2017" @default.
- W2617361225 type Work @default.
- W2617361225 sameAs 2617361225 @default.
- W2617361225 citedByCount "3" @default.
- W2617361225 countsByYear W26173612252018 @default.
- W2617361225 crossrefType "journal-article" @default.
- W2617361225 hasAuthorship W2617361225A5051371429 @default.
- W2617361225 hasAuthorship W2617361225A5062627428 @default.
- W2617361225 hasAuthorship W2617361225A5087619194 @default.
- W2617361225 hasConcept C11413529 @default.
- W2617361225 hasConcept C118552586 @default.
- W2617361225 hasConcept C119857082 @default.
- W2617361225 hasConcept C153180895 @default.
- W2617361225 hasConcept C154945302 @default.
- W2617361225 hasConcept C157286648 @default.
- W2617361225 hasConcept C41008148 @default.
- W2617361225 hasConcept C48044578 @default.
- W2617361225 hasConcept C522805319 @default.
- W2617361225 hasConcept C71924100 @default.
- W2617361225 hasConcept C77088390 @default.
- W2617361225 hasConceptScore W2617361225C11413529 @default.
- W2617361225 hasConceptScore W2617361225C118552586 @default.
- W2617361225 hasConceptScore W2617361225C119857082 @default.
- W2617361225 hasConceptScore W2617361225C153180895 @default.
- W2617361225 hasConceptScore W2617361225C154945302 @default.
- W2617361225 hasConceptScore W2617361225C157286648 @default.
- W2617361225 hasConceptScore W2617361225C41008148 @default.
- W2617361225 hasConceptScore W2617361225C48044578 @default.
- W2617361225 hasConceptScore W2617361225C522805319 @default.
- W2617361225 hasConceptScore W2617361225C71924100 @default.
- W2617361225 hasConceptScore W2617361225C77088390 @default.
- W2617361225 hasLocation W26173612251 @default.
- W2617361225 hasLocation W26173612252 @default.
- W2617361225 hasOpenAccess W2617361225 @default.
- W2617361225 hasPrimaryLocation W26173612251 @default.
- W2617361225 hasRelatedWork W1525643724 @default.
- W2617361225 hasRelatedWork W1527726406 @default.
- W2617361225 hasRelatedWork W2067938758 @default.
- W2617361225 hasRelatedWork W2302028273 @default.
- W2617361225 hasRelatedWork W2364921833 @default.
- W2617361225 hasRelatedWork W2382623646 @default.
- W2617361225 hasRelatedWork W2388030554 @default.
- W2617361225 hasRelatedWork W2961085424 @default.
- W2617361225 hasRelatedWork W3087771547 @default.
- W2617361225 hasRelatedWork W4313203779 @default.
- W2617361225 hasVolume "158" @default.
- W2617361225 isParatext "false" @default.
- W2617361225 isRetracted "false" @default.
- W2617361225 magId "2617361225" @default.
- W2617361225 workType "article" @default.