生物医学工程学杂志

生物医学工程学杂志

基于改进遗传算法优化反向传播神经网络的癫痫发作检测方法分析

查看全文

为了提高计算机化癫痫发作检测的准确性和检测效率,本文提出了一种基于改进遗传算法的优化反向传播(IGA-BP)神经网络的癫痫诊断方法,以期利用该方法可以实现临床癫痫病症的快速、高效检测。该方法首先对癫痫脑电信号进行线性与非线性相结合的特征提取,通过高斯混合模型(GMM)对癫痫特征聚簇集合分析,利用最大期望(EM)算法估算高斯混合模型参量,获取遗传算法(GA)选择算子的最优参数组合,实现对遗传算法的改进,用改进的遗传算法调整反向传播(BP)神经网络以获取最佳初始权值和阈值,建立改进遗传算法优化的 BP 神经网络模型。利用该模型对癫痫脑电信号分类识别,最终实现癫痫病症的自动检测。与传统遗传算法优化的 BP(GA-BP)神经网络相比较,本文所提出的方法提高了种群的收敛速度、减小了分类误差,在癫痫病症自动检测中提高了检测准确率并缩短了检测时间,在临床癫痫发作诊断中具有重要的应用价值。

In order to improve the accuracy and efficiency of automatic seizure detection, the paper proposes a method based on improved genetic algorithm optimization back propagation (IGA-BP) neural network for epilepsy diagnosis, and uses the method to achieve detection of clinical epilepsy rapidly and effectively. Firstly, the method extracted the linear and nonlinear features of the epileptic electroencephalogram (EEG) signals and used a Gaussian mixture model (GMM) to perform cluster analysis on EEG features. Next, expectation maximization (EM) algorithm was used to estimate GMM parameters to calculate the optimal parameters for the selection operator of genetic algorithm (GA). The initial weights and thresholds of the BP neural network were obtained through using the improved genetic algorithm. Finally, the optimized BP neural network is used for the classification of the epileptic EEG signals to detect the epileptic seizure automatically. Compared with the traditional genetic algorithm optimization back propagation (GA-BP), the IGA-BP neural network can improve the population convergence rate and reduce the classification error. In the process of automatic detection of epilepsy, the method improves the detection accuracy in the automatic detection of epilepsy disorders and reduced inspection time. It has important application value in the clinical diagnosis and treatment of epilepsy.

关键词: 癫痫发作检测; 遗传算法; BP 神经网络; 选择算子; EM 算法

Key words: epileptic seizure detection; genetic algorithm; BP neural networks; selection operator; EM algorithm

引用本文: 刘光达, 魏星, 张尚, 蔡靖, 刘颂阳. 基于改进遗传算法优化反向传播神经网络的癫痫发作检测方法分析. 生物医学工程学杂志, 2019, 36(1): 24-32. doi: 10.7507/1001-5515.201806039 复制

登录后 ,请手动点击刷新查看全文内容。 没有账号,
登录后 ,请手动点击刷新查看图表内容。 没有账号,
1. Harikumar R, Sukanesh R, Bharathi P A. Genetic algorithm optimization of fuzzy outputs for classification of epilepsy risk levels from EEG signals// Conference Record of the Thirty-Eighth Asilomar Conference on Signals, Systems and Computers, 2004, 2: 1585-1589.
2. Marchena P, Díaz M, Esteller R, et al. Use of artificial neural networks for classification intracraneal EEG signals from epileptic patients.World Congress on Medical Physics and Biomedical Engineering, Munich, Germany. Springer Berlin Heidelberg, 2009: 398-401.
3. 李伟, 何鹏举, 杨恒, 等. 基于粗糙集和改进遗传算法优化 BP 神经网络的算法研究. 西北工业大学学报, 2012, 30(4): 601-606.
4. 彭基伟, 吕文华, 行鸿彦, 等. 基于改进 GA-BP 神经网络的湿度传感器的温度补偿. 仪器仪表学报, 2013, 34(1): 153-160.
5. Yang Yongqiang, Wang Guijun, Yang Yang. Parameters optimization of polygonal fuzzy neural networks based on GA-BP hybrid algorithm. International Journal of Machine Learning and Cybernetics, 2014, 5(5): 815-822.
6. He Fei, Zhang Lingying. Mold breakout prediction in slab continuous casting based on combined method of GA-BP neural network and logic rules. International Journal of Advanced Manufacturing Technology, 2018, 95(9/12): 4081-4089.
7. Dhiman R, Saini J S, Priyanka. Genetic algorithms tuned expert model for detection of epileptic seizures from EEG signatures. Appl Soft Comput, 2014, 19(2): 8-17.
8. Ding Jinliang, Zhao Liang, Liu Changxin, et al. GA-based principal component selection for production performance estimation in mineral processing. Computers & Electrical Engineering, 2014, 40(5): 1447-1459.
9. 孟庆芳, 陈珊珊, 陈月辉, 等. 基于递归量化分析与支持向量机的癫痫脑电自动检测方法. 物理学报, 2014, 63(5): 050501-050506.
10. 韩凌, 王宏. 基于空频域特征分析方法的癫痫发作预测. 仪器仪表学报, 2014, 35(11): 2501-2507.
11. 刘浩然, 赵翠香, 李轩, 等. 一种基于改进遗传算法的神经网络优化算法研究. 仪器仪表学报, 2016, 37(7): 1573-1580.
12. Thistleton W J, Marsh J A, Nelson K, et al. Generalized Box-MUller method for generating formula not shown-Gaussian random deviates. IEEE Transactions on Information Theory It, 2007, 53(6): 4805-4810.
13. Zablotskiy S, Pitakrat T, Zablotskaya K, et al. GMM parameter estimation by means of EM and genetic algorithms. Lecture Notes in Computer Science, 2011, 6761: 527-536.
14. Liu Guangquan, Huang Gan, Meng Jianjun, et al. Improved GMM with parameter initialization for unsupervised adaptation of brain-computer interface. Int J Numer Method Biomed Eng, 2010, 26(6): 681-691.
15. Watanabe H, Muramatsu S, Kikuchi H. Interval calculation of EM algorithm for GMM parameter estimation//IEEE International Symposium on Circuits and Systems, IEEE, 2010: 2686-2689.
16. Trad D, Al-Ani T, Jemni M. A feature extraction technique of EEG based on EMD-BP for motor imagery classification in BCI//2015 5th International Conference on Information & Communication Technology and Accessibility (ICTA), IEEE, 2015: 1-6.
17. 袁琦, 周卫东, 李淑芳, 等. 基于 ELM 和近似熵的脑电信号检测方法. 仪器仪表学报, 2012, 33(3): 514-519.
18. Yan X, Sun X, Yin Q. Multipara meter sensitivity analysis of operational energy efficiency for inland river ships based on back propagation neural network method. Marine Technology Society Journal, 2015, 49(1): 148-153.