Automatic epileptic seizure detection method based on spatio-temporal feature fusion.

Comput Methods Biomech Biomed Engin

School of Intelligent Manufacturing, Longdong University, Qingyang, Gansu, P. R. China.

Published: September 2025


Category Ranking

98%

Total Visits

921

Avg Visit Duration

2 minutes

Citations

20

Article Abstract

This paper proposes a spatiotemporal feature fusion method for automatic epileptic seizure detection, integrating Common Spatial Pattern (CSP) and Least Squares Support Vector Machine (LSSVM). First, it reconstructs electroencephalogram (EEG) noise using Ensemble Empirical Mode Decomposition (EEMD), then decomposes the original EEG signals using improved EEMD (IEEMD). Next, features are extracted from temporal and spatial dimensions to form a feature set. The classification process adopts a novel dual-classification mode based on LSSVM ultimately achieving high-performance automatic recognition of normal, seizure, and interictal EEG signals. Validated on Bonn and CHB-MIT EEG datasets, the IEEMD algorithm achieves 99.57% ± 0.02 accuracy on Bonn and 96.43% overall accuracy on CHB-MIT. Results show IEEMD and spatiotemporal features effectively address low interictal-ictal recognition rates in existing studies, offering a reliable means for epileptic seizure prediction.

Download full-text PDF

Source
http://dx.doi.org/10.1080/10255842.2025.2551845DOI Listing

Publication Analysis

Top Keywords

epileptic seizure
12
automatic epileptic
8
seizure detection
8
feature fusion
8
eeg signals
8
seizure
4
detection method
4
method based
4
based spatio-temporal
4
spatio-temporal feature
4

Similar Publications

Background And Purpose: Neuroinflammation is increasingly recognised to contribute to drug-resistant epilepsy. Activation of ATP-gated P2X7 receptors has emerged as an important upstream mechanism, and increased P2X7 receptor expression is present in the seizure focus in rodent models and patients. Pharmacological antagonists of P2X7 receptors attenuate seizures in rodents, but this has not been explored in human neural networks.

View Article and Find Full Text PDF

Fahr's syndrome is a rare neurological condition marked by unusual calcifications in the basal ganglia and other brain regions, often resulting from metabolic disorders, such as hypoparathyroidism. Secondary hypoparathyroidism, a frequent complication of total thyroidectomy, can lead to Fahr's syndrome, manifesting as movement disorders, seizures, psychiatric symptoms and indications of calcium deficiency. This case report discusses a woman in her mid-30s who developed Fahr's syndrome due to secondary hypoparathyroidism after total thyroidectomy.

View Article and Find Full Text PDF

Regulation of neurogenesis and neuronal migration by Rrm2 and Timp3 following seizures.

Neurobiol Dis

September 2025

Department of Neuroscience, Developmental and Regenerative Biology, The University of Texas at San Antonio, San Antonio, TX, USA; Brain Health Consortium, The University of Texas at San Antonio, San Antonio, TX, USA. Electronic address:

Temporal lobe epilepsy is associated with aberrant neurogenesis and ectopic migration of adult-born granule cells (abGCs), yet the molecular mechanisms driving these changes remain poorly defined. Using a pilocarpine-induced mouse model of temporal lobe epilepsy and chemogenetic silencing of abGCs via Designer Receptors Exclusively Activated by Designer Drugs (DREADDs), we previously demonstrated that abGC inhibition reduces both ectopic migration and seizure susceptibility. To identify underlying molecular regulators, we performed RNA sequencing of FACS-isolated abGCs and identified Rrm2 and Timp3 as top candidate genes modulated by seizure activity and neuronal silencing.

View Article and Find Full Text PDF

UBQLN4 regulates seizures by promoting the proteasomal degradation of GluN2B.

Neurobiol Dis

September 2025

Department of Neurology, the First Affiliated Hospital of Chongqing Medical University, Chongqing Key Laboratory of Major Neurological and Mental Disorders, Chongqing Key Laboratory of Neurology, Chongqing, China. Electronic address:

Ubiquilin 4 (UBQLN4) is an important molecule that regulates protein degradation through the ubiquitin-proteasome pathway. This study found that UBQLN4 expression is significantly reduced in a chronic epilepsy mouse model induced by kainic acid, primarily localized in neurons and widely distributed at excitatory post-synapses. Experiments involving adeno-associated virus-mediated overexpression or knockdown of UBQLN4 indicate that a reduction in UBQLN4 increases susceptibility to and severity of epilepsy, while its overexpression has a protective effect.

View Article and Find Full Text PDF

The effect of recurrent seizures on the gradual deterioration of the white matter structural network and the potential molecular mechanisms that underlie the baseline and longitudinal changes in network topology in temporal lobe epilepsy (TLE) remain unclear. Therefore, we used diffusion tensor imaging (DTI) scans and neuropsychiatric assessments for 28 patients with unilateral TLE at baseline and follow-up, and for 28 healthy controls (HC). The topological properties of the structural network were calculated using graph theoretical analyses.

View Article and Find Full Text PDF