Dynamic time warping and machine learning for signal quality assessment of pulsatile signals

被引:195
作者
Li, Q. [1 ,2 ]
Clifford, G. D. [2 ]
机构
[1] Shandong Univ, Sch Med, Inst Biomed Engn, Jinan 250012, Shandong, Peoples R China
[2] Univ Oxford, Dept Engn Sci, Inst Biomed Engn, Oxford OX1 3PJ, England
关键词
artificial neural network; dynamic time warping; machine learning; multi-layer perceptron; photoplethysmograph; pulsatile signal; signal quality assessment; PULSE OXIMETRY; MOTION ARTIFACT; PHOTOPLETHYSMOGRAPHY; SEGMENTATION; REDUCTION;
D O I
10.1088/0967-3334/33/9/1491
中图分类号
Q6 [生物物理学];
学科分类号
071011 ;
摘要
In this work, we describe a beat-by-beat method for assessing the clinical utility of pulsatile waveforms, primarily recorded from cardiovascular blood volume or pressure changes, concentrating on the photoplethysmogram (PPG). Physiological blood flow is nonstationary, with pulses changing in height, width and morphology due to changes in heart rate, cardiac output, sensor type and hardware or software pre-processing requirements. Moreover, considerable inter-individual and sensor-location variability exists. Simple template matching methods are therefore inappropriate, and a patient-specific adaptive initialization is therefore required. We introduce dynamic time warping to stretch each beat to match a running template and combine it with several other features related to signal quality, including correlation and the percentage of the beat that appeared to be clipped. The features were then presented to a multi-layer perceptron neural network to learn the relationships between the parameters in the presence of good- and bad-quality pulses. An expert-labeled database of 1055 segments of PPG, each 6 s long, recorded from 104 separate critical care admissions during both normal and verified arrhythmic events, was used to train and test our algorithms. An accuracy of 97.5% on the training set and 95.2% on test set was found. The algorithm could be deployed as a stand-alone signal quality assessment algorithm for vetting the clinical utility of PPG traces or any similar quasi-periodic signal.
引用
收藏
页码:1491 / 1501
页数:11
相关论文
共 29 条
[1]  
Addison P S, 2010, Signal processing techniques for determining signal quality using a wavelet transform ratio surface, Patent No. [US Patent No. 2010/0298728, 20100298728]
[2]   Photoplethysmography and its application in clinical physiological measurement [J].
Allen, John .
PHYSIOLOGICAL MEASUREMENT, 2007, 28 (03) :R1-R39
[3]  
Chan K. W., 2002, Proceedings of IEEE Sensors 2002. First IEEE International Conference on Sensors (Cat. No.02CH37394), P1343, DOI 10.1109/ICSENS.2002.1037314
[4]  
Clifford GD, 2011, COMPUT CARDIOL CONF, V38, P285
[5]  
Clifford G. D., 2002, THESIS OXFORD U OXFO
[6]  
Deshmane A. V., 2009, THESIS MIT CAMBRIDGE
[7]   PTT Variability for Discrimination of Sleep Apnea Related Decreases in the Amplitude Fluctuations of PPG Signal in Children [J].
Gil, Eduardo ;
Bailon, Raquel ;
Maria Vergara, Jose ;
Laguna, Pablo .
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 2010, 57 (05) :1079-1088
[8]   PhysioBank, PhysioToolkit, and PhysioNet - Components of a new research resource for complex physiologic signals [J].
Goldberger, AL ;
Amaral, LAN ;
Glass, L ;
Hausdorff, JM ;
Ivanov, PC ;
Mark, RG ;
Mietus, JE ;
Moody, GB ;
Peng, CK ;
Stanley, HE .
CIRCULATION, 2000, 101 (23) :E215-E220
[9]  
Graybeal JM, 2004, P ANN INT IEEE EMBS, V26, P5363
[10]   A new method for pulse oximetry possessing inherent insensitivity to artifact [J].
Hayes, MJ ;
Smith, PR .
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 2001, 48 (04) :452-461