应用遗传算法优化神经网络的致死性心电节律辨识算法研究

发布时间：2018-02-24 01:22

本文关键词： 心脏骤停自动体外除颤仪遗传算法反向传播神经网络　出处：《生物医学工程学杂志》2017年03期 　论文类型：期刊论文

【摘要】：致死性心电节律的辨识和分类是自动体外除颤仪的关键任务。本文对已存在的心电节律辨识算法提取出的21个特征值进行了回顾性研究,并基于这些特征值构建了一个遗传算法优化的反向传播神经网络。以数据库提供的1 343例心电信号样本用于实验。实验结果表明,本文构建的神经网络在对窦性节律、心室颤动、室性心动过速、心脏停搏4类心电信号的辨识分类上有很好的表现,在测试集上的平衡准确性高达99.06%;相较已存在的算法,辨识性能更好。将该算法应用在自动体外除颤仪上,将进一步提高除颤前节律分析的可靠性,最终提高心脏骤停的存活率。
[Abstract]:The identification and classification of fatal ECG rhythm is the key task of automatic external defibrillator. In this paper, 21 eigenvalues extracted from existing ECG rhythm identification algorithms are studied retrospectively. Based on these eigenvalues, a genetic algorithm optimized back propagation neural network is constructed. 1 343 ECG samples provided by the database are used in the experiment. The experimental results show that the neural network constructed in this paper is in the sinus rhythm. Ventricular fibrillation (VF), ventricular tachycardia (VT), cardioplegia (CPB) have a good performance in recognition and classification of electrocardiogram (ECG) signals, and the accuracy of balancing on the test set is as high as 99.06%. The application of the algorithm to the automatic external defibrillator will further improve the reliability of the analysis of the defibrillation rhythm and ultimately improve the survival rate of cardiac arrest.
【作者单位】：军事医学科学院卫生装备研究所;
【基金】：国家自然科学基金资助项目(81501551)
【分类号】：R541.7;TP18

【相似文献】