Liu Mingze, Liu Jie, Xu Mengna, Liu Yasheng, Li Jie, Nie Weiwei, Yuan Qi
Shandong Province Key Laboratory of Medical Physics and Image Processing Technology, School of Physics and Electronics, Shandong Normal University, Jinan, 250358, China.
Department of Pediatric Intensive Care Unit, Shandong Provincial Maternal and Child Health Care Hospital Affiliated to Qingdao University, Jinan, 250014, China.
Sci Rep. 2025 Mar 28;15(1):10755. doi: 10.1038/s41598-025-88270-3.
Despite two decades of extensive research into electroencephalogram (EEG)-based automated seizure detection analysis, the persistent imbalance between seizure and non-seizure categories remains a significant challenge. This study integrated meta-sampling with an ensemble classifier to address the issue of imbalanced classification existing in seizure detection. In this framework, a meta-sampler was employed to autonomously acquire undersampling strategies from EEG data. During each iteration, the meta-sampler interacted with the external environment on a single occasion with the objective of deriving an optimal sampling strategy through this interactive learning process. It was anticipated that optimal sampling strategies would be derived through interactive learning. And then the soft Actor-Critic algorithm was employed to address the non-differentiable optimization issue associated with the meta-sampler. Consequently, this framework adaptively selected training EEG data, and learned effective cascaded integrated classifiers from unbalanced epileptic EEG data. Besides, the time domain, nonlinear and entropy-based EEG features were extracted from five frequency bands (δ, θ, α, β, and γ) and were selected by Semi-JMI to be fed into this imbalanced classification framework. The proposed detection system achieved a sensitivity of 92.58%, a specificity of 92.51%, and an accuracy of 92.52% on the scalp EEG dataset. On the intracranial EEG dataset, the average sensitivity, specificity, and accuracy were 98.56%, 98.82%, and 98.7%, respectively. The experimental comparisons demonstrated that the system outperformed other state-of-the-art methods, and showed robustness in the face of label corruption.
尽管对基于脑电图（EEG）的自动癫痫发作检测分析进行了二十年的广泛研究，但癫痫发作和非癫痫发作类别之间持续存在的不平衡仍然是一个重大挑战。本研究将元采样与集成分类器相结合，以解决癫痫发作检测中存在的不平衡分类问题。在此框架中，采用元采样器从EEG数据中自主获取欠采样策略。在每次迭代中，元采样器与外部环境进行一次交互，目的是通过这种交互式学习过程得出最优采样策略。预计将通过交互式学习得出最优采样策略。然后采用软演员-评论家算法来解决与元采样器相关的不可微优化问题。因此，该框架自适应地选择训练EEG数据，并从不平衡的癫痫EEG数据中学习有效的级联集成分类器。此外，从五个频段（δ、θ、α、β和γ）提取基于时域、非线性和熵的EEG特征，并通过半联合互信息（Semi-JMI）进行选择，以输入到这个不平衡分类框架中。所提出的检测系统在头皮EEG数据集上的灵敏度为92.58%，特异性为92.51%，准确率为92.52%。在颅内EEG数据集上，平均灵敏度、特异性和准确率分别为98.56%、98.82%和98.7%。实验比较表明，该系统优于其他现有方法，并且在面对标签损坏时表现出鲁棒性。
