Virtual resection evaluation based on sEEG propagation network for drug-resistant epilepsy DOI Creative Commons
Jie Sun, Yan Niu,

Yanqing Dong

и другие.

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Окт. 26, 2024

Drug-resistant epilepsy with frequent seizures are considered to undergo surgery become seizure-free, but seizure-free rates have not dramatically improved, partly due imprecise intervention locations. To address this clinical need, we construct effective connectivity reveal brain dynamics. Based on the propagation path captured by high order connectivity, calculate control centrality evaluation scheme of excised area. We used three datasets: simulation dataset, and public dataset. The epileptogenic network was quantified calculating high-order connection obtain accurate path, based this, combined outdegree index for virtual resection. By removing electrodes recalculating centrality, quantify each electrode or region's evaluate resection scheme. Three datasets obtained consistent results. track find obvious inflection points occurring during excision process. minimum targets were comparing different schemes without recurrence. data multiple found that after resection, reaches a stable state is less likely continue spreading. quantitative analysis possible scheme, finally best area epilepsy, which assist in developing surgical plans.

Язык: Английский

Deep Learning for Epileptic Seizure Detection Using a Causal-Spatio-Temporal Model Based on Transfer Entropy DOI Creative Commons
Jie Sun, Jie Xiang,

Yanqing Dong

и другие.

Entropy, Год журнала: 2024, Номер 26(10), С. 853 - 853

Опубликована: Окт. 10, 2024

Drug-resistant epilepsy is frequent, persistent, and brings a heavy economic burden to patients their families. Traditional detection methods ignore the causal relationship of seizures focus on single time or spatial dimension, effect varies greatly in different patients. Therefore, it necessary research accurate automatic technology We propose causal-spatio-temporal graph attention network (CSTGAT), which uses transfer entropy (TE) construct between multiple channels, combining (GAT) bi-directional long short-term memory (BiLSTM) capture temporal dynamic correlation topological structure information. The accuracy, specificity, sensitivity SWEZ dataset were 97.24%, 97.92%, 98.11%. accuracy private reached 98.55%. effectiveness each module was proven through ablation experiments impact construction compared. experimental results indicate that constructed by TE could accurately information flow epileptic seizures, GAT BiLSTM spatiotemporal correlations. This model captures relationships correlations two datasets, overcomes variability patients, may contribute clinical surgical planning.

Язык: Английский

Процитировано

1
Virtual resection evaluation based on sEEG propagation network for drug-resistant epilepsy DOI Creative Commons
Jie Sun, Yan Niu,

Yanqing Dong

и другие.

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Окт. 26, 2024

Drug-resistant epilepsy with frequent seizures are considered to undergo surgery become seizure-free, but seizure-free rates have not dramatically improved, partly due imprecise intervention locations. To address this clinical need, we construct effective connectivity reveal brain dynamics. Based on the propagation path captured by high order connectivity, calculate control centrality evaluation scheme of excised area. We used three datasets: simulation dataset, and public dataset. The epileptogenic network was quantified calculating high-order connection obtain accurate path, based this, combined outdegree index for virtual resection. By removing electrodes recalculating centrality, quantify each electrode or region's evaluate resection scheme. Three datasets obtained consistent results. track find obvious inflection points occurring during excision process. minimum targets were comparing different schemes without recurrence. data multiple found that after resection, reaches a stable state is less likely continue spreading. quantitative analysis possible scheme, finally best area epilepsy, which assist in developing surgical plans.

Язык: Английский

Процитировано

0