[1]任灵,黄玉丹,陈颖.基于交叉对比神经网络的心音分类[J].中国医学物理学杂志,2021,38(10):1251-1257.[doi:DOI:10.3969/j.issn.1005-202X.2021.10.012]
　REN Ling,HUANG Yudan,CHEN Ying.Heart sound classification based on cross-contrast neural network[J].Chinese Journal of Medical Physics,2021,38(10):1251-1257.[doi:DOI:10.3969/j.issn.1005-202X.2021.10.012]
点击复制

基于交叉对比神经网络的心音分类()
分享到：

《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]

卷:
38卷
期数:
2021年第10期
页码:
1251-1257
栏目:
医学信号处理与医学仪器
出版日期:
2021-10-27

文章信息/Info

Title:
Heart sound classification based on cross-contrast neural network
文章编号:
1005-202X（2021）10-1251-07
作者:
任灵; 黄玉丹; 陈颖
南京大学电子科学与工程学院， 江苏 南京 210023
Author(s):
REN Ling;  HUANG Yudan;  CHEN Ying
School of Electrical Science and Engineering, Nanjing University, Nanjing 210023, China
关键词:
心音分类; 交叉对比神经网络; 基于信息的相似度度量理论; 深度学习
Keywords:
Keywords: heart sound classification cross-contrast neural network information-based similarity measurement theory deep learning
分类号:
R318.5
DOI:
DOI:10.3969/j.issn.1005-202X.2021.10.012
文献标志码:
A
摘要:
目的：通过交叉对比神经网络（CCNN）实现心音信号的自动分类，从而对心血管疾病进行早期诊断。方法：实验基于PhysioNet/Cinc 2016心音数据库。训练集和测试集数据来自互斥的健康受试者/病理患者，并以4：1的比例进行划分，输入CCNN。CCNN利用深度卷积神经网络进行特征提取，结合基于信息的相似度度量理论（IBS），对特征向量间的相似性进行度量并分类。结果：实验结果得出灵敏度为0.834 6，特异性为0.962 3，最终大赛综合得分为0.898 5。结论：CCNN使用交叉对比的输入模式扩充数据量，引入信号间的对比信息，同时在神经网络的训练过程中应用统计学思想，使网络具备良好的泛化性，更加适应医学数据量较少的场景，在心音分类中取得较好的结果。
Abstract:
Abstract: Objective To automatically classify heart sound signals by cross-contrast neural network (CCNN), thereby realizing the early diagnosis of cardiovascular diseases. Methods The experiment was carried out based on PhysioNet/Cinc 2016 heart sound database. The training set and test set data which came from mutually exclusive healthy subjects/pathological patients were divided at a ratio of 4:1 and then input into CCNN. Finally, CCNN used deep convolutional neural network for feature extraction and was combined with information-based similarity measurement theory to measure and classify the similarity between feature vectors. Results The sensitivity and specificity of CCNN for heart sound classification in the experiment were 0.834 6 and 0.962 3, respectively, and the overall score reached 0.898 5. Conclusion By expanding the amount of data using a cross-contrast input mode, introducing contrast information between signals and applying statistical ideas in the training process of neural networks, CCNN has good generalization and is more suitable for small medical data, having a good performance in heart sound classification.

相似文献/References:

[1]马晶,蔡文杰,杨利.基于机器学习的心音识别分类研究[J].中国医学物理学杂志,2021,38(1):75.[doi:DOI:10.3969/j.issn.1005-202X.2021.01.013]
　MA Jing,CAI Wenjie,YANG Li.Research on heart sounds classification based on machine learning[J].Chinese Journal of Medical Physics,2021,38(10):75.[doi:DOI:10.3969/j.issn.1005-202X.2021.01.013]
[2]冯帅,刘飞飞,伍昕宇,等.基于最小均方误差对数谱幅度估计的心音降噪算法[J].中国医学物理学杂志,2023,40(11):1370.[doi:DOI:10.3969/j.issn.1005-202X.2023.11.009]
　FENG Shuai,LIU Feifei,WU Xinyu,et al.Heart sound denoising algorithm based on minimum mean-square error log-spectral amplitude estimation[J].Chinese Journal of Medical Physics,2023,40(10):1370.[doi:DOI:10.3969/j.issn.1005-202X.2023.11.009]

备注/Memo

备注/Memo:
【收稿日期】2021-04-13 【基金项目】国家自然科学基金（81671777）；江苏省重点研发计划（BE2017679, BE2016733） 【作者简介】任灵，硕士研究生，研究方向：生物医学工程，E-mail: MF1823042@smail.nju.edu.cn 【通信作者】陈颖，副教授，研究生导师，研究方向：生物医学工程，E-mail: yingchen@nju.edu.cn
更新日期/Last Update: 2021-10-29