Unsupervised anomaly detection in peripheral venous pressure signals with hidden Markov models

Unsupervised anomaly detection in peripheral venous pressure signals with hidden Markov models
复制标题

使用隐马尔可夫模型进行外周静脉压力信号的无监督异常检测

DOI:
10.1016/j.bspc.2020.102126
复制
发表时间:
2020
影响因子:
5.1
通讯作者:
Jensen, Morten O.
Jensen, Morten O.
中科院分区:
工程技术2区
文献类型:
--
作者:
Hayat, Md Abul;Wu, Jingxian;Bonasso, Patrick C.;Sexton, Kevin W.;Jensen, Hanna K.;Dassinger, Melvin S.;Jensen, Morten O.

文献摘要

参考文献

相似文献

本文提出了一种外周静脉压(PVP)信号的自动异常检测和消除算法,可用于预测人体血管内容量损失。 PVP 信号采集是一种微创手术,可以使用标准外周静脉 (PIV) 导管和商用压力监测传感器来执行。 PVP 信号非常容易受到运动和噪声伪影的影响,例如患者移动或 PIV 线的意外操作。 PVP 信号中的异常可能会破坏有用信息并严重影响 PVP 信号分析的完整性。我们建议通过利用 PVP 信号的特性来检测和消除此类异常。具体来说,带有卡尔曼滤波器的动态线性模型 (DLM) 用于跟踪和预测 PVP 信号的时域演化。然后使用隐马尔可夫模型 (HMM) 对卡尔曼滤波器的预测残差进行建模,并使用隐马尔可夫链的二进制状态对信号的正常和异常状态进行建模。 HMM 参数以及隐藏状态是通过使用无监督学习算法和修改后的 Baum-Welch 方法迭代估计的。异常检测算法应用于 24 名患有肥厚性幽门狭窄的儿科患者的临床数据。实验结果表明,所提出的无监督异常检测算法可以有效地去除 PVP 信号中的异常,而不需要训练阶段。该算法还可以应用于其他时间序列信号,例如心电图(ECG)和光电容积描记图(PPG)信号。
This paper proposes an automatic anomaly detection and removal algorithm for peripheral venous pressure (PVP) signals, which can be used to predict intravascular volume loss in humans. PVP signal collection is a minimally invasive procedure that can be performed by using a standard peripheral intravenous (PIV) catheter and a commercial pressure-monitoring transducer. PVP signals are highly susceptible to motion and noise artifacts such as patient movements or unintended manipulation of PIV lines. Anomalies in PVP signals can corrupt useful information and seriously affect the integrity of PVP signal analysis. We propose to detect and remove such anomalies by exploiting the properties of PVP signals. Specifically, a dynamic linear model (DLM) with a Kalman filter is used to track and predict the time-domain evolution of PVP signals. The prediction residuals of the Kalman filter are then modeled with a hidden Markov model (HMM), with the normal and anomalous status of the signal modeled by using binary states of a hidden Markov chain. The HMM parameters along with the hidden states are iteratively estimated by using an unsupervised learning algorithm with a modified Baum–Welch method. The anomaly detection algorithm is applied to clinical data from a cohort of 24 pediatric patients with hypertrophic pyloric stenosis. Experimental results demonstrate that the proposed unsupervised anomaly detection algorithm can efficiently remove anomalies in PVP signals without the need of a training phase. The algorithm can also be applied to other time series signals, such as Electrocardiography (ECG) and Photoplethysmogram (PPG) signals.
DOI: 10.1016/j.transproceed.2005.12.057
发表时间: 2006-03-01
影响因子: 0.9
作者:
Hadimioglu, N;Ertug, Z;Demirbas, A
通讯作者: Demirbas, A
DOI: 10.1007/s00134-003-2052-0
发表时间: 2004-04-01
影响因子: 38.9
作者:
Desjardins, R;Denault, AY;Martineau, R
通讯作者: Martineau, R
外周静脉压作为原位肝移植期间中心静脉压的预测因子。
DOI: 10.1016/j.jclinane.2005.09.031
发表时间: 2006
影响因子: 6.7
作者:
N. Hoftman;M. Braunfeld;Gil D. Hoftman;A. Mahajan
通讯作者: A. Mahajan
DOI: 10.1111/j.2517-6161.1977.tb01600.x
发表时间: 1977-01-01
期刊: JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-METHODOLOGICAL
影响因子: --
作者:
DEMPSTER, AP;LAIRD, NM;RUBIN, DB
通讯作者: RUBIN, DB
静脉生理学预测婴儿麻醉引起的低血压
DOI: 10.1016/j.jamcollsurg.2018.08.313
发表时间: 2018
影响因子: 5.2
作者:
P. Bonasso;K. Sexton;A. Hayat;A. Al;Jingxian Wu;H. Jensen;M. Jensen;Samuel D. Smith;Jeffrey M. Burford;M. Dassinger
通讯作者: M. Dassinger