|本期目录/Table of Contents|

[1]许逸钰,周涛,邹玉涛,等.基于数据增强、黏菌算法与坐标注意力机制的癫痫发作亚型分类方法[J].浙江理工大学学报,2026,55-56(自科三):325-323.
　XU Yiyu,ZHOU Tao,ZOU Yutao,et al.Epilepsy seizure subtype classification based on data augmentation, slime mold algorithm, and coordinate attention mechanism[J].Journal of Zhejiang Sci-Tech University,2026,55-56(自科三):325-323.
点击复制

基于数据增强、黏菌算法与坐标注意力机制的癫痫发作亚型分类方法()

浙江理工大学学报[ISSN:1673-3851/CN:33-1338/TS]

卷:
55-56
期数:
2026年自科第三期
页码:
325-323
栏目:
出版日期:
2026-05-10

文章信息/Info

Title:
Epilepsy seizure subtype classification based on data augmentation, slime mold algorithm, and coordinate attention mechanism
文章编号:
1673-3851(2026) 05-0315-09
作者:
许逸钰 ; 周涛 ; 邹玉涛 ; 姜楠 ; 蒋路茸 ; 曹天傲
浙江理工大学 ,a. 信息科学与工程学院(网络空间安全学院);b. 肾虚瘀浊证研究与转化重点实验室 ,杭州 31001
Author(s):
XU Yiyu; ZHOU Tao ; ZOU Yutao ; JIANG Nan ; JIANG Lurong ; CAO Tianao
a. School of Information Science and Engineering(School of Cyber Scienceand Technology) ; b. Key Laboratory for Research and Translation of Kidney Deficiency-Stasis-Turbidity Disease, Zhejiang Sci-Tech University, Hangzhou 310018, China
关键词:
癫痫发作亚型分类 ; 去噪扩散概率模型 ; 黏菌算法 ; 坐标注意力机制 ; 脑电图
分类号:
TP391
文献标志码:
A
摘要:
癫痫发作亚型的精准分类对诊疗有重要意义 ,但目前仍存在类别分布不均衡、特征冗余度较高等关键问题 ,为此设计了一种基于去噪扩散概率模型(Denoising diffusion probabilistic model, DDPM) 、黏菌算法(Slime moldalgorithm, SMA)与坐标注意力(Coordinateattention, CA)机制的深度学习方法 。首先 ,利用 DDPM生成高保真少数类脑电样本 , 以缓解原始数据的分布不平衡问题 。其次 ,使用SMA算法对特征进行自适应筛选降维 , 降低特征冗余并增强关键特征的表达能力 。再次 ,建立融合 CA机制的双向长短期记忆网络(Bi-directionallong short-term memory,BiLSTM) , 以提高多通道脑电信号的时空依赖建模能力 。最后 ,采用 TUSZ数据集 ,对原始脑电信号实施预处理及样本划分 ,并根据 DDPM数据增强、SMA筛选特征、CA-BiLSTM模型训练以及性能评估等 4 个方面展开实验 。 实验结果表明 ,该方法在 6类癫痫发作亚型分类任务上获得了 96.54%的平均准确率与 0.9687的平均 F1分数 ,在模型稳定性和鲁棒性方面优于常见方法 。该方法能够在复杂临床脑电数据条件下提升癫痫发作亚型分类的性能 ,为临床精准诊断与个体化治疗提供一定的技术支持。

参考文献/References:

[1] Ioannou P, Foster D L, Sander J W, et al. The burden of epilepsy and unmet need in people with focal seizures[J]. Brain and Behavior, 2022, 12(9): e2589.

[2] 张思琪, 杨添淞, 马帅, 等. 深度学习在脑卒中诊断与防治中的研究进展[J]. 磁共振成像, 2022, 13(11): 125-128.
[3] 中国抗癫痫协会. 临床诊疗指南: 癫痫病分册: 2023修订版[M]. 北京: 人民卫生出版社, 2023: 48-69.
[4] 丁晶, 汪昕. 癫痫诊疗指南解读[J]. 临床内科杂志, 2016, 33(2): 142-144.
[5] Fraiwan L, Alkhodari M. Classification of focal and non-focal epileptic patients using single channel EEG and long short-term memory learning system[J]. IEEE Access, 2020, 8: 77255-77262.
[6] Feng Y M, Zheng R Z, Cui X N, et al. 3D residual-attention deep-network-based childhood epilepsy syndrome classification[J]. Knowledge-Based Systems, 2022, 248: 108856.
[7] Cao J W, Chen Y H, Zheng R Z, et al. DSMN-ESS: dual-stream multitask network for epilepsy syndrome classification and seizure detection[J]. IEEE Transactions on Instrumentation and Measurement, 2023, 72: 4010212.
[8] Ahmad I, Yao C, Li L, et al. An efficient feature selection and explainable classification method for EEG-based epileptic seizure detection[J]. Journal of Information Security and Applications, 2024, 80: 103654.
[9] He H B, Garcia E A. Learning from imbalanced data[J]. IEEE Transactions on Knowledge and Data Engineering, 2009, 21(9): 1263-1284.
[10] Abd Elrahman S M, Abraham A. A review of class imbalance problem[J]. Journal of Network and Innovative Computing, 2013, 1: 332-340.
[11] 王夙喆, 张雪英, 陈晓玉, 等. 基于有效注意力和GAN结合的脑卒中EEG增强算法[J]. 计算机工程, 2024, 50(8): 336-344.
[12] 曹崴, 李奇. 基于Transformer与SNN的WGAN-GP癫痫脑电数据增强方法[J]. 现代电子技术, 2025, 48(23): 167-175.
[13] Nichol A Q, Dhariwal P. Improved denoising diffusion probabilistic models[C]//Proceedings of the 38th International Conference on Machine Learning. PMLR, 2021: 8162-8171.
[14] Dhariwal P, Nichol A. Diffusion models beat GANs on image synthesis[C]//Proceedings of the 35th International Conference on Neural Information Processing Systems. ACM, 2021: 8780-8794.
[15] Yi H, Hou L, Jin Y, et al. Time series diffusion method: a denoising diffusion probabilistic model for vibration signal generation[J]. Mechanical Systems and Signal Processing, 2024, 216: 111481.
[16] Zhao Z X, Bai H W, Zhu Y Z, et al. DDFM: Denoising diffusion model for multi-modality image fusion[C]//2023 IEEE/CVF International Conference on Computer Vision (ICCV). Paris, France: IEEE, 2023: 8048-8059.
[17] Gupta G, Khan S, Guleria V, et al. DDPM: a dengue disease prediction and diagnosis model using sentiment analysis and machine learning algorithms[J]. Diagnostics, 2023, 13(6): 1093.
[18] Obeid I, Picone J. The temple university hospital EEG data corpus[J]. Frontiers in Neuroscience, 2016, 10: 196.
[19] Ho J, Jain A N, Abbeel P. Denoising diffusion probabilistic models[C]//Larochelle H, Ranzato M, Hadsell R, et al. Advances in Neural Information Processing Systems. Vol.33. RedHook, NY: Curran Associates, Inc., 2020: 6840-6851.
[20] Wu D, Li J, Dong F, et al. Classification of seizure types based on multi-class specific bands common spatial pattern and penalized ensemble model[J]. Biomedical Signal Processing and Control, 2023, 79: 104118.
[21] 王剑, 成婷, 宋政阳, 等. 基于深度学习的癫痫异常信号检测和分类模型[J]. 电子测量技术, 2025, 48(17): 113-124.
[22] Tuncer E, Bolat E D. Channel based epilepsy seizure type detection from electroencephalography (EEG) signals with machine learning techniques[J]. Biocybernetics and Biomedical Engineering, 2022, 42(2): 575-595.
[23] Albaqami H, Hassan G M, Datta A. Wavelet-based multi-class seizure type classification system[J]. Applied Sciences, 2022, 12(11): 5702.
[24] 陈沛宇, 叶佩华, 黄辉. 基于三维卷积神经网络的多频段癫痫发作类型分类[J]. 工业控制计算机, 2022, 35(12): 108-109.
[25] 吴端坡, 励杰, 应娜. 基于二维脑网络特征选择的癫痫发作类型分类算法研究[J]. 实验室研究与探索, 2023, 42(12): 73-78.
[26] Jia G, Lam H K, Althoefer K. Variable weight algorithm for convolutional neural networks and its applications to classification of seizure phases and types[J]. Pattern Recognition, 2022, 121: 108226.
[27] Hu Y, Liu J, Sun R, et al. Classification of epileptic seizures in EEG data based on iterative gated graph convolution network[J]. Frontiers in Computational Neuroscience, 2024, 18: 1454529.
[28] Huang H, Chen P, Wen J, et al. Multiband seizure type classification based on 3D convolution with attention mechanisms[J]. Computers in Biology and Medicine, 2023, 166: 107517.

备注/Memo

备注/Memo:
基金项目 : 浙江省自然科学基金青年基金项目(25222260-D) ;浙江省教育厅一般科研项目(Y202457174)收稿日期 : 2025-12-13 网络出版日期 : 2026-03-05
作者简介 : 许逸钰(2000— ) ,男 ,江苏苏州人 ,硕士研究生 ,主要从事脑电信号方面的研究。通信作者: 蒋路茸,E-mail:jianglurong@zstu. edu. cn
更新日期/Last Update: 2026-05-07