一种基于心率和深层学习的心电图分类算法
发布时间：2018-11-20 15:12
【摘要】：目的研究适用于远程医疗服务系统、体检中心和临床应用的心电图(electrocardiogram,ECG)正异常分类算法。方法首先,通过心率筛除异常数据。然后,对于心率判为正常的心电图,采用LCNN对心电图再次进行正异常分类,并对多个LCNN的分类结果进行融合。结果在15万多条记录的临床数据集上测试,取得了84.77%的准确率,85.19%的灵敏度和84.45%的特异性。结论该实验结果优于对照文献,同时对应用于远程医疗和体检中心的计算辅助分析方法具有一定的参考价值。
[Abstract]:Objective to study the positive classification algorithm of electrocardiogram (electrocardiogram,ECG) for telemedicine service system, physical examination center and clinical application. Methods first, abnormal data were screened by heart rate. Then, for the ECG with normal heart rate, LCNN was used to classify the ECG again, and the classification results of multiple LCNN were fused. Results the accuracy, sensitivity and specificity were 84.77%, 85.19% and 84.45% respectively. Conclusion the results of the experiment are superior to those of the control literature, and it has some reference value for the Computer-Aided Analysis method used in telemedicine and physical examination center.
【作者单位】： 上海大学通信与信息工程学院;中国科学院苏州纳米技术与纳米仿生研究所;
【分类号】：TP18;R540.41
，
本文编号：2345244
[Abstract]:Objective to study the positive classification algorithm of electrocardiogram (electrocardiogram,ECG) for telemedicine service system, physical examination center and clinical application. Methods first, abnormal data were screened by heart rate. Then, for the ECG with normal heart rate, LCNN was used to classify the ECG again, and the classification results of multiple LCNN were fused. Results the accuracy, sensitivity and specificity were 84.77%, 85.19% and 84.45% respectively. Conclusion the results of the experiment are superior to those of the control literature, and it has some reference value for the Computer-Aided Analysis method used in telemedicine and physical examination center.
【作者单位】： 上海大学通信与信息工程学院;中国科学院苏州纳米技术与纳米仿生研究所;
【分类号】：TP18;R540.41
，
本文编号：2345244
本文链接：https://www.wllwen.com/huliyixuelunwen/2345244.html
最近更新
教材专著
