Author(s): Peng, H (Peng, Hong); Lei, C (Lei, Chang); Zheng, SZ (Zheng, Shuzhen); Zhao, CJ (Zhao, Chengjian); Wu, CY (Wu, Chunyun); Sun, JQ (Sun, Jieqiong); Hu, B (Hu, Bin)

Source: COMPUTERS IN BIOLOGY AND MEDICINE Volume: 132 Article Number: 104338 DOI: 10.1016/j.compbiomed.2021.104338 Published: MAY 2021

Abstract: Epileptic seizure detection is of great significance in the diagnosis of epilepsy and relieving the heavy workload of visual inspection of electroencephalogram (EEG) recordings. This paper presents a novel method for seizure detection using the Stein kernel-based sparse representation (SR) for EEG recordings. Different from the traditional SR scheme that works with vector data in Euclidean space, the Stein kernel-based SR framework is constructed for seizure detection in the space of the symmetric positive definite (SPD) matrices, which form a Riemannian manifold. Due to the non-Euclidean geometry of the Riemannian manifold, the Stein kernel on the manifold permits the embedding of the manifold in a high-dimensional reproducing kernel Hilbert space (RKHS) to perform SR. In the Stein kernel-based SR framework, EEG samples are described by SPD matrices in the form of covariance descriptors (CovDs). Then, a test EEG sample is sparsely represented on the training set, and the test sample is classified as a member of the class, which leads to the minimum reconstructed residual. Finally, by using three widely used EEG datasets to evaluate the detection performance of the proposed method, the experimental results demonstrate that it achieves good classification accuracy on each dataset. Furthermore, the fast computational speed of the Stein kernel-based SR also meets the basic requirements for real-time seizure detection.

Accession Number: WOS:000649715900002

PubMed ID: 33780870

Author Identifiers:

Author        Web of Science ResearcherID        ORCID Number

Lei, Chang         AAO-2920-2021

ISSN: 0010-4825

eISSN: 1879-0534

Full Text: https://www.sciencedirect.com/science/article/pii/S0010482521001323?via%3Dihub