没有合适的资源?快使用搜索试试~ 我知道了~
资源推荐
资源详情
资源评论
摘 要
I
摘 要
目前,心血管疾病已经成为人类身体健康的头号杀手,心律不齐是心血管疾病的
一种典型代表,对心律不齐准确分类和检测具有十分重要的意义。心电图因其成本低
廉和良好的无创性,已逐渐成为医生诊断心律不齐疾病所用最多的依据。随着人们的
健康意识的提高和家用心电测试仪的逐渐普及,心电图的数量急剧增加,有限的医疗
资源无法完成海量的心电数据诊断,因此心律不齐自动分类技术已变得至关重要。
传统机器学习方法依赖手工特征设计,很难将心电信号特征充分挖掘;心电信号
存在较大的个体差异性,模型的泛化能力不强。以上因素都限制了心律不齐分类技术
的现实应用。针对以上问题,本文将集成学习和半监督学习的方法运用到心律不齐分
类技术上,面向病人间通用和特定病人专用两种不同分类模式,本文分别做了进一步
深入研究,论文主要工作如下:
(1)面向病人间心律不齐分类,提出了一种基于集成深度学习的心律不齐分类算
法。该方法使用双向长短期记忆网络和改进的迁移卷积神经网络分别提取心电信号的
时域、频域、几何和状态变迁特征。该方法充分挖掘心电信号特征,通过集成所提取
的多种特征对心电信号进行联合分类,有利于提升模型的准确率。考虑到不同模型对
最终的分类结果影响不同,使用了一种加权集成的策略,根据模型在训练时的表现为
其赋予不同的集成权重,充分且合理地利用了心电信号各种特征。经 MIT-BIH 心律不
齐数据库验证,本文所提算法的总体准确率达到了 96.2%。
(2)面向特定病人心律不齐分类,提出一种融合集成学习和半监督学习的心律不
齐分类算法。在通用数据上加入小部分特定病人的专用数据,通过半监督训练的方法
得到最后的特定模型,该方法能够学习到特定病人的心电特征,提升模型泛化性能。
为了减少半监督训练过程中的伪标签噪声,使用一种融合网络不确定性和预测概率的
方法筛选伪标签,增强了伪标签数据的可靠性。最后将各个模型集成,进一步增强了
模型的稳定性。通过 MIT-BIH 心律不齐数据库数据验证,本文算法达到了 98.1%的总
体准确率。
关键词 心律不齐 集成学习 半监督学习 深度学习
河北大学硕士学位论文
II
Abstract
At present, cardiovascular disease has become the leading cause of death. As a typical
cardiovascular disease, the accurate classification and detection of arrhythmia is of great
significance. Electrocardiogram has become the most common basis for doctors to diagnose
arrhythmia due to the non-invasive and low-cost. The number of electrocardiogram has
increased sharply with the improvement of health awareness and the universal of home
electrocardiograph. However, the limited medical resources cannot complete the massive
diagnosis of ECG. Therefore, the automatic classification for arrhythmia has become a crucial
technology.
The ECG features cannot be mined completely since the traditional machine learning
methods rely on the design of manual features. The large individual differences limit the
generalization of the model. The above factors limit the practical application of arrhythmia
classification technology. To solve the problems mentioned above, the integrated learning and
semi-supervised learning are introduced to the classification of arrhythmia. This paper makes
further research on the two different modes of arrhythmia classification. The main work is as
follows:
(1) An arrhythmia classification algorithm based on ensemble deep learning is proposed
to perform the inter-patient arrhythmia classification. The features are extracted from time
domain, frequency domain, geometric and State transition of ECG respectively by bi-
directional long Short-Term memory network and improved transfer convolutional neural
network. The features are fully extracted and the multi features are integrated to perform the
joint classification of ECG, which is beneficial to improve the accuracy of the model. A
weighted integration strategy is adopted considering that the final classification results vary
with different models. The features of ECG are used fully and reasonably by assigning
different weights to models according to the performance of during training. The proposed
model is tested on the arrhythmia database of MIT-BIH and the overall accuracy has reached
96.2%.
(2) An arrhythmia classification algorithm based on the combination of ensemble
learning and semi-supervised learning is proposed to perform patient-specific arrhythmia
Abstract
III
classification. The training data consisted of common data and a few special data comes from
specific patient. The final specific model was obtained through semi-supervised training,
which can learn the ECG features of a specific patient to improve the generalization of the
model. The combination of network uncertainty and prediction probability was introduced to
reduce the pseudo-label noise in the process of semi-supervised training, which enhanced the
accuracy of pseudo-labels. The various models are integrated to further enhance the stability
of the model. The proposed method was tested on the arrhythmia database of MIT-BIH and
the overall accuracy had reached 98.1%.
Keywords Arrhythmia Ensemble learning Semi-supervised learning Deep learning
河北大学硕士学位论文
IV
目 录
第一章 绪论 .............................................................................................................................. 1
1.1 研究背景及意义 .......................................................................................................... 1
1.2 研究现状 ...................................................................................................................... 2
1.2.1 基于传统机器学习的心律不齐分类研究现状 ............................................... 2
1.2.2 基于深度学习的心律不齐分类研究现状 ....................................................... 3
1.2.3 心律不齐分类技术存在的问题 ....................................................................... 4
1.3 本文研究工作和结构安排 .......................................................................................... 5
第二章 相关技术介绍 .............................................................................................................. 7
2.1 心电信号相关知识 ...................................................................................................... 7
2.2 心律不齐相关知识 ...................................................................................................... 8
2.2.1 心律不齐疾病简介 ............................................................................................ 8
2.2.2 MIT-BIH 心律不齐数据库和 AAMI 标准介绍 .............................................. 8
2.2.3 心律不齐分类模式 ......................................................................................... 10
2.3 神经网络基础 ............................................................................................................ 10
2.3.1 神经元和多层感知器 ..................................................................................... 10
2.3.2 卷积神经网络 ................................................................................................. 12
2.3.3 Xception 网络结构 .......................................................................................... 14
2.4 迁移学习和集成学习 ................................................................................................ 16
2.5 本章小结 .................................................................................................................... 17
第三章 基于集成深度学习的病人间心律不齐分类 ............................................................ 18
3.1 引言 ............................................................................................................................ 18
3.2 基于集成深度学习的病人间心律不齐分类算法 .................................................... 18
3.2.1 算法框架 .......................................................................................................... 18
目 录
V
3.2.2 一维心电信号转二维图像 ............................................................................. 19
3.2.3 网络模型 ......................................................................................................... 22
3.2.4 加权集成策略 ................................................................................................. 27
3.3 实验与分析 ................................................................................................................ 30
3.3.1 数据来源及预处理 ......................................................................................... 30
3.3.2 实验设置与评价标准 ..................................................................................... 31
3.3.3 实验结果分析 ................................................................................................. 32
3.4 本章小结 .................................................................................................................... 36
第四章 融合集成学习和半监督学习的特定病人心律不齐分类 ........................................ 37
4.1 引言 ............................................................................................................................ 37
4.2 融合集成学习和半监督学习的特定病人心律不齐分类算法 ................................ 37
4.2.1 算法框架 ......................................................................................................... 37
4.2.2 模型训练与伪标签制作 ................................................................................. 38
4.3 实验设置及评价标准 ................................................................................................ 41
4.3.1 数据来源及预处理 ......................................................................................... 41
4.3.2 实验设置与评价标准 ...................................................................................... 42
4.4 实验结果及其分析 .................................................................................................... 42
4.5 本章小结 .................................................................................................................... 46
第五章 总结与展望 ................................................................................................................ 47
5.1 工作总结 .................................................................................................................... 47
5.2 工作展望 .................................................................................................................... 48
致 谢 ........................................................................................................................................ 54
攻读学位期间取得的成果 ...................................................................................................... 55
剩余56页未读,继续阅读
资源评论
icwx_7550592
- 粉丝: 17
- 资源: 7163
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
最新资源
- Idea Note.apk
- 通过navicat操作数据库.flv
- frameworks 之线程类使用 原生和自定义的线程类
- 华为OD机试C卷- 园区参观路径(Java & JS & Python & C).md-私信看全套OD代码及解析
- 华为OD机试C卷- 寻找最优的路测线路(Java & JS & Python & C).md-私信看全套OD代码及解析
- 华为OD机试C卷- 田忌赛马(Java & JS & Python & C).md-私信看全套OD代码及解析
- 华为OD机试C卷- 特殊的加密算法(Java & JS & Python & C).md-私信看全套OD代码及解析
- 华为OD机试C卷- 贪吃蛇(Java & JS & Python).md-私信看全套OD代码及解析
- 电子迎宾器和声光控照明开关multsim仿真和课设报告
- 深度学习 KAN 最新发布论文
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功