人工神经网络是由大量并行工作的神经元组成的智能仿生模型,它在模式识别领域已经展示出了广阔的应用前景.鉴于单一心率变异性(HRV)指标所表达出来的信息具有片面性,很难用一个单一的指标来完全分类充盈性心衰(CHF)患者和健康人的不足.本研究提出联合HRV信号分析的时域、频域、非线性方法,选取多个指标作为诊断CHF的特征参数,以BP神经网络为分类器实现对充盈性心衰的诊断.经过10 000次的训练、验证与仿真测试,该网络模型对于全样本集的识别正确率最优高达99.14％,平均可达86.97％.结果表明:联合线性(时域、频域)以及非线性分析方法可以更全面地揭示心脏的动力学特征,从而提高充盈性心衰的诊断正确率.%Information delivered by a single heart rate variability ( HRV) index is inadequate and it is very difficult to completely classify the congestive heart failure (CHF) patients from healthy people by using a single index. Artificial neural network, inspired by biological nervous systems, is composed of simple elements operating in parallel and open up very broad vistas in the field of pattern recognition. Based on composing time-domain analysis, frequency-domain analysis and nonlinear analysis of HRV signals, several indices were extracted as the feature parameters for the diagnosis of CHF. Then BP network was trained and applied to the diagnosis of congestive heart failure. After 10000 times of training, validating and testing, the optimal recognition rate achieved 99. 14% with 86. 97% on average. The results suggest that the conjunctive application of linear and nonlinear methods of HRV can extract more information underlying the complex dynamic systems.
展开▼
