面向临床心电图分析的深层学习算法

被引:34
作者
金林鹏 [1 ,2 ]
董军 [1 ]
机构
[1] 中国科学院苏州纳米技术与纳米仿生研究所
[2] 中国科学院大学
关键词
深层学习; 心电图(ECG); 非线性拟合; 卷积神经网络; 数据分布与分类;
D O I
暂无
中图分类号
TP391.4 [模式识别与装置];
学科分类号
0811 ; 081101 ; 081104 ; 1405 ;
摘要
作为模式识别研究的典型应用,计算机辅助心电图分析在可穿戴终端及面向基层的"云"服务平台具有重要价值.本文首先说明了面向临床的心电图分类模型的复杂性,接着从非线性函数拟合能力角度分析了现有的特征提取算法和分类算法,并采用深层学习构造心电图分类模型.针对多导联心电图这种特殊的二维结构,提出导联卷积神经网络,并利用"平移起始点"和"加噪"增加训练样本数.通过15万多条记录的测试数据,取得了准确率为83.66%和AUC为0.9086的成绩.最后我们还移植分类模型到移动终端设备,其实时分析结果满足应用需求.
引用
收藏
页码:398 / 416
页数:19
相关论文
共 13 条
[1]  
融合领域知识的心电图分类方法研究[D]. 王丽苹.华东师范大学 2013
[2]   An R-peak detection method based on peaks of Shannon energy envelope [J].
Zhu, Honghai ;
Dong, Jun .
BIOMEDICAL SIGNAL PROCESSING AND CONTROL, 2013, 8 (05) :466-474
[3]   Computer-based rhythm diagnosis and its possible influence on nonexpert electrocardiogram readers [J].
Hakacova, Nina ;
Tragardh-Johansson, Elin ;
Wagner, Galen S. ;
Maynard, Charles ;
Pahlm, Olle .
JOURNAL OF ELECTROCARDIOLOGY, 2012, 45 (01) :18-22
[4]   An asymmetric classifier based on partial least squares [J].
Qu, Hai-Ni ;
Li, Guo-Zheng ;
Xu, Wei-Sheng .
PATTERN RECOGNITION, 2010, 43 (10) :3448-3457
[5]  
A multi-stage automatic arrhythmia recognition and classification system[J] . Yakup Kutlu,Damla Kuntalp.Computers in Biology and Medicine . 2010 (1)
[6]   A two-stage mechanism for registration and classification of ECG using Gaussian mixture model [J].
Martis, Roshan Joy ;
Chakraborty, Chandan ;
Ray, Ajoy K. .
PATTERN RECOGNITION, 2009, 42 (11) :2979-2988
[7]  
Combining recurrent neural networks with eigenvector methods for classification of ECG beats[J] . Elif Derya übeyli.Digital Signal Processing . 2008 (2)
[8]   Recurrent neural networks with composite features for detection of electrocardiographic changes in partial epileptic patients [J].
Ubeyli, Elif Derya .
COMPUTERS IN BIOLOGY AND MEDICINE, 2008, 38 (03) :401-410
[9]   Verification of humans using the electrocardiogram [J].
Wuebbeler, Gerd ;
Stavridis, Manuel ;
Kreiseler, Dieter ;
Bousseljot, Ralf-Dieter ;
Elster, Clemens .
PATTERN RECOGNITION LETTERS, 2007, 28 (10) :1172-1175
[10]  
A trainable feature extractor for handwritten digit recognition[J] . Fabien Lauer,Ching Y. Suen,Gérard Bloch.Pattern Recognition . 2006 (6)