摘要
针对心肌梗死疾病的快速准确定位,提出一种结合基因算法和极限学习机(GA-ELM)的新定位算法。对多导联心电图(electrocardiogram,ECG)进行去噪预处理;分别定位出每个R波峰的位置,并将心电信号分割成相互独立的心搏序列,作为混合模型的输入;采用(Physikalisch-Technische Bundesanstalt,PTB)心肌梗死数据库验证所提算法的有效性。利用十折交叉验证法评测算法定位的精度,实验结果表明,GA-ELM混合模型对心肌梗死定位的准确率达到98.42%,性能优于其他文献的算法,对医护工作者确定心肌梗死部位具有重要的临床意义。
Aiming at the rapid and accurate localization of myocardial infarction diseases,this paper proposes a new localization algorithm combining genetic algorithm and extreme learning machine(GA-ELM).The denoising preprocessing of the multi-lead electrocardiogram(ECG)was performed,then the position of each R peak was located separately and the ECG signal was segmented into independent heartbeat sequences as the input of the hybrid model.We used PTB myocardial infarction database to verify the effectiveness of the proposed algorithm.The accuracy of algorithm was evaluated by the 10-fold cross-validation method.The experimental results show that the accuracy of GA-ELM hybrid model is 98.42%,which is superior to other algorithms.It has important clinical significance for medical workers to determine the location of myocardial infarction.
作者
张行进
李润川
张宏坡
逯鹏
王宗敏
Zhang Xingjin;Li Runchuan;Zhang Hongpo;Lu Peng;Wang Zongmin(State Key Laboratory of Mathematical Engineering and Advanced Computing,People s Liberation Army of China Information Engineering University,Zhengzhou 450001,Henan,China;Cooperative Innovation Center of Internet Healthcare,Zhengzhou University,Zhengzhou 450000,Henan,China;Research Institute of Industrial Technology,Zhengzhou University,Zhengzhou 450000,Henan,China)
出处
《计算机应用与软件》
北大核心
2020年第1期186-191,共6页
Computer Applications and Software
基金
国家重点研发计划项目(2017YFB1401200)
兵团重点领域科技攻关计划项目(2018AB017)
河南省高等学校重点科研项目(18A520055)
河南省科技攻关项目(142102210069)
