多特征信息融合LSTM-RNN检测OSA方法

朱兆坤; 李金宝

doi:10.7544/issn1000-1239.2020.20190583

多特征信息融合LSTM-RNN检测OSA方法

朱兆坤^1,2,3,
李金宝¹

¹(齐鲁工业大学(山东省科学院)山东省人工智能研究院济南 250014)
²(黑龙江大学计算机科学技术学院哈尔滨 150080)
³(黑龙江大学软件学院哈尔滨 150080) (zzklove3344@hotmail.com)

基金项目: 国家自然科学基金项目(61370222)；黑龙江省自然科学基金重点项目(ZD2019F003)

详细信息

中图分类号: TP391
计量
- 文章访问数: 1099
- HTML全文浏览量: 3
- PDF下载量: 400
出版历程
- 发布日期: 2020-11-30

Multi-Feature Information Fusion LSTM-RNN Detection for OSA

Zhu Zhaokun^1,2,3,
Li Jinbao¹

¹(Shandong Artificial Intelligence Institute, Qilu University of Technology (Shandong Academy of Science), Jinan 250014)
²(School of Computer Science and Tecnology, Heilongjiang University, Harbin 150080)
³(Software Technology Institute, Heilongjiang University, Harbin 150080)

Funds: This work was supported by the National Natural Science Foundation of China (61370222) and the Key Program of the Natural Science Foundation of Heilongjiang Province of China (ZD2019F003).

摘要

摘要: 阻塞性睡眠呼吸暂停(obstructive sleep apnea, OSA)是最常见的睡眠呼吸疾病，它对人体的很多生理系统尤其对心血管系统是一个潜在的威胁.现有使用心电信号(electrocardiograph, ECG)提取浅层特征检测OSA的方法在长片段、高噪声的ECG信号和大数据集上表现较差.针对上述问题，提出一种多特征心电信号融合的长短期记忆循环神经网络，融合从ECG信号中提取的多种浅层特征信号，通过在融合信号上学习深层特征来检测OSA，提升模型在长片段ECG上的检测准确率和大数据集上的泛化能力.同时还针对浅层特征信号提出一种有效的数据预处理方法，用以突出OSA的时序变化，提高神经网络训练的收敛性，并降低由异常值噪声带来的影响，进一步提升模型在高噪声ECG片段上的检测准确率.实验证明：提出的方法在片段OSA检测准确率上优于已有的方法.
- 阻塞性睡眠呼吸暂停 /
- 心电信号 /
- 预处理 /
- 长短时记忆 /
- 循环神经网络
Abstract: Obstructive sleep apnea (OSA) is the most common sleep respiratory disorder, and it is a potential threat to many physiological systems, especially the cardiovascular system. Most of the previous methods for OSA detection extracted the shallow features from electrocardiograph (ECG) which would be used in classifiers, and they failed to achieve excellent performances on the ECG signal with high noise and large datasets. To solve this kind of problem, this paper proposes a long short-term memory recurrent neural network (LSTM-RNN) based on combination of multiple kinds of feature signals. The method fuses multiple kinds of shallow feature signals that are extracted from ECG signals and learns the deep feature from the fused signals. The accuracy of OSA detection model in long ECG segments is increased and the generalization ability on large datasets is improved. An effective preprocessing method is propesed for shallow feature signals to highlight the variation of OSA time sequences. The preprocessing method may improve the convergence of training neural networks, reduce the impact of outlier noise, and further improve the detection accuracy of the model for the ECG segments with high noise. The experimental results indicate that our method is superior to the existing methods in the accuracy of per-segment OSA detection.
- obstructive sleep apnea /
- electrocardiograph /
- preprocessing /
- long short-term memory /
- recurrent neural network

HTML全文

参考文献(0)

施引文献(24)

期刊类型引用(8)

1.	张宇姣，徐健，吴迪. 基于图表示学习的知识图谱时序推理模型. 济南大学学报(自然科学版). 2025(02): 272-277 . 百度学术
2.	陆佳炜，王小定，朱昊天，程振波，肖刚. 一种融合实体图上下文的三维旋转知识图谱表示学习. 小型微型计算机系统. 2023(01): 124-131 . 百度学术
3.	陈小英，熊盛武，王盛，张士伟. 基于上下文时序关联的时序知识图谱嵌入方法. 武汉大学学报(理学版). 2023(02): 249-257 . 百度学术
4.	卢菁，陈婉璐，刘丛. KGU-SP:一种挖掘标准模式的知识图谱更新方法. 小型微型计算机系统. 2023(06): 1177-1183 . 百度学术
5.	魏飞鸣，许倩倩，顾网平，李永晨. 知识图谱在探测与识别领域中的应用分析. 制导与引信. 2023(04): 1-8+28 . 百度学术
6.	马昂，于艳华，杨胜利，石川，李劼，蔡修秀. 基于强化学习的知识图谱综述. 计算机研究与发展. 2022(08): 1694-1722 . 本站查看
7.	宁原隆，周刚，卢记仓，杨大伟，张田. 一种融合关系路径与实体描述信息的知识图谱表示学习方法. 计算机研究与发展. 2022(09): 1966-1979 . 本站查看
8.	夏毅，兰明敬，陈晓慧，罗军勇，周刚，何鹏. 可解释的知识图谱推理方法综述. 网络与信息安全学报. 2022(05): 1-25 . 百度学术