基于双向长短时记忆网络和自注意力机制的心音分类
作者:
作者单位:

1.南京邮电大学通信与信息工程学院,南京 210003;2.智能信息处理与通信技术省高校重点实验室(南京邮电大学),南京 210003;3.南京邮电大学计算机学院,南京 210023;4.南京医科大学附属儿童医院心胸外科,南京 210008

基金项目:

国家自然科学基金(72074038);江苏省卫生健康委员会重点项目(K2023036);南京市留学人员科技创新项目(NJKCZYZZ2023-04);南京邮电大学引进人才科研启动基金(NY223030)。


Heart Sound Classification Using Bi-LSTM and Self-attention Mechanism
Author:
Affiliation:

1.School of Communications and Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210003, China;2.Jiangsu Key Laboratory of Intelligent Information Processing and Communication Technology, Nanjing University of Posts and Telecommunications, Nanjing 210003, China;3.School of Computer Science, Nanjing University of Posts and Telecommunications, Nanjing 210023, China;4.Department of Cardiothoracic Surgery, Children’s Hospital of Nanjing Medical University, Nanjing 210008, China

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
    摘要:

    心音听诊是早期筛查心脏病的有效诊断方法。为了提高异常心音检测性能,提出了一种基于双向长短时记忆(Bi-directional long short-term memory,Bi-LSTM)网络和自注意力机制(Self-attention mechanism,SA)的心音分类算法。对心音信号进行分帧处理,提取每帧心音信号的梅尔频率倒谱系数(Mel-frequency cepstral coefficients,MFCC)特征;将MFCC特征序列输入Bi-LSTM网络,利用Bi-LSTM网络提取心音信号的时域上下文特征;通过自注意力机制动态调整Bi-LSTM网络各时间步输出特征的权重,得到有利于分类的更具鉴别性的心音特征;通过Softmax分类器实现正常/异常心音的分类。在PhysioNet/CinC Challenge 2016心音数据集上对所提出的算法使用10折交叉验证法进行了评估,得到0.942 5的灵敏度、0.943 7的特异度、0.836 7的精度、0.886 5的F1得分和0.943 4的准确率,优于对比的典型算法。实验结果表明,该算法在无需进行心音分段的基础上就能有效实现异常心音检测,具有潜在的临床应用前景。

    Abstract:

    Heart sound auscultation is an effective diagnostic method for early screening of heart disease. In order to improve the performance of abnormal heart sound detection, this paper proposes a heart sound classification algorithm based on bi-directional long short-term memory (Bi-LSTM) network and self-attention mechanism (SA). Firstly, the heart sound signal is partitioned into frames, and the Mel-frequency cepstral coefficients (MFCC) features are extracted from each frame of the heart sound signal. Next, the MFCC feature sequence is input into the Bi-LSTM network to extract the temporal contextual features of the heart sound signals. Then, the weights of the features output from the Bi-LSTM network at each time step are dynamically adjusted through self-attention mechanism, and more discriminative heart sound features that are conducive to classification are obtained. Finally, the Softmax classifier is used to classify normal/abnormal heart sounds. The proposed algorithm is evaluated using 10-fold cross-validation on the heart sound dataset provided by PhysioNet/CinC Challenge 2016, and achieves sensitivity of 0.942 5, specificity of 0.943 7, accuracy of 0.836 7, F1 score of 0.886 5, and accuracy of 0.943 4, respectively, which are superior to typical comparative algorithms. Experimental results show that the proposed algorithm can effectively detect abnormal heart sounds without the need for heart sound segmentation, and has potential clinical application prospects.

    参考文献
    相似文献
    引证文献
引用本文

卢官明,李齐健,卢峻禾,戚继荣,赵宇航,王洋,魏金生.基于双向长短时记忆网络和自注意力机制的心音分类[J].数据采集与处理,2025,40(2):456-468

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
历史
  • 收稿日期:2024-03-30
  • 最后修改日期:2024-04-29
  • 在线发布日期: 2025-04-11