## Signal Processing Methods for Heart Rate Variability (1998)

Citations: | 3 - 2 self |

### BibTeX

@TECHREPORT{College98signalprocessing,

author = {St. Cross College and Supervised Prof and L. Tarassenko and Gari D. Clifford and Gari D. Clifford},

title = {Signal Processing Methods for Heart Rate Variability},

institution = {},

year = {1998}

}

### OpenURL

### Abstract

and except where otherwise stated, describes my own research.

### Citations

4885 |
Neural networks for pattern recognition
- Bishop
- 2000
(Show Context)
Citation Context ...is defined over all training patterns, Ô, in the training set as � � � � Ô � �� � � Û���� 77 � � Û��� Ô �sØ Ô � (3.6) (3.7)sThe squared error, �, can be minimised using the method of gradient descent =-=[34]-=-. This requires the gradient to be calculated with respect to each weight, Û�� and Û��. The weight update equations for the hidden and output layers are given as follows: Û � �� Û � �� � Û � ��s� � Û�... |

2159 | E.: A new approach to linear filtering and prediction problems - Kalman - 1960 |

1094 | A Practical Guide to Splines - Boor |

312 |
On the use of windows for harmonic analysis with the discrete Fourier transform
- Harris
- 1978
(Show Context)
Citation Context ...s the form of side-lobes in the periodogram. This spectral leakage is classically minimised through windowing or other mathematically equivalent techniques. In Harris’ detailed paper on DFT windowing =-=[108]-=- he catalogued many of the most useful windowing 110sfunctions and their side-lobe structure. He showed that a Hamming window given by [35] Ï Ø� � ���s��� Ó×��Ø�℄ � � � � � ���� Æs� (4.24) provides th... |

262 |
A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages
- Rechtschaffen, Kales
- 1968
(Show Context)
Citation Context ... on how to measure levels of mental activity while a subject is awake because of extraneous factors [118, 209], there is a generally agreed classification of mental activity for subjects during sleep =-=[166, 231]-=- (see section 5.4.1). During sleep, the influence of extraneous factors is reduced. Recent reports have shown that significant differences in HRV metrics exist between some sleep states and in particu... |

230 |
The use of fast fourier transform for the estimation of power spectra: A method based on time averaging over short modi¯ed periodograms
- Welch
- 1967
(Show Context)
Citation Context ...nt of Æ Ì � �� on the number of points and hence to a lower limit on Æ of 240 beats (an average heart rate of 48 bpm if all beats in a 5-minute segment are used). 2 A windowed version of equation 4.6 =-=[279]-=-. 111s4.3.2 Sampling frequency Most of the papers in the field of HRV report on the use of re-sampling rates between and �ÀÞ (although 1Hz and 10Hz have also been used) [114, 181, 184] to produce an e... |

199 | Multidimensional independent component analysis
- Cardoso
- 1998
(Show Context)
Citation Context ... analysis to be employed more universally. Recently, blind source separation, or independent component analysis (ICA), has been used to separate biomedical signals into their (independent) components =-=[22, 42, 135, 278, 283]-=-. ICA involves taking multiple observations of a signal which is a mixture of several statistically independent signals and identifying the independent signal sources (usually by minimising or maximis... |

117 |
A real-time QRS detection algorithm
- Pan, Tompkins
- 1985
(Show Context)
Citation Context ...ost-processing decision rules that operate on this output to identify relevant sections within the ECG for detailed analysis in order to extract the R-peak which is referred to as the fiducial point (=-=[215]-=-). It is this point that corresponds to the clinicians’ labelling in the annotation files. This first stage is broken down into four separate distinct filtering procedures: 1. 5-15Hz Band Pass Filteri... |

73 |
Estimation and tracking: principles, techniques and software, Artech House
- Bar-Shalom, Li
- 1993
(Show Context)
Citation Context ...hermore, prior information about the noise and signal structure of 187sthe data being analysed can be included in the estimation. It may also be possible to use the Extended Kalman Filter formulation =-=[23]-=- to quantify the nonlinear components of an RR tachogram. These new (unevenly sampled) techniques could be compared to the linear stationary techniques explored in this thesis to assess whether quanti... |

69 | Improved Surrogate Data for Nonlinearity Tests - Schreiber, Schmitz - 1996 |

68 | Pattern Recognition and Neural Networks (Cambridge - Ripley - 1996 |

55 |
Updating the singular value decomposition
- BUNCH, NIELSEN
- 1978
(Show Context)
Citation Context ... Ø � �s�� ��� (3.33) Equations (3.29), (3.30) and (3.31) represent an analytical solution to determine the weights of the auto-associative MLP ‘in one pass’ with as few as Æ� �Æ� Ç Æ� multiplications =-=[39]-=-. With � ���this compares favourably with the usual error back-propagation training algorithm which requires �Æ� � Ç �Æ�� Ç �Æ � � multiplications, where � is the number of iterations through the Æ ¢�... |

48 | Kalman filtering: theory and practice using MATLAB. 3rd - Grewal, Andrews - 2008 |

37 |
Quantitative Investigation of QRS Detection Rules Using the MIT/BIH Arrhythmia Database
- Hamilton, Tompkins
(Show Context)
Citation Context ...gh pass filtering or other filtering methods then they offer the best system for QRS detection. A detailed study of an algorithm which incorporates these features (the method of Hamilton and Tompkins =-=[107]-=-) is presented later in this chapter. It should be noted that the Hamilton and Tompkins algorithm, like all of the non-pattern matching algorithms, is essentially a peak energy-amplitude detection alg... |

32 | A Dynamical Model for Generating Synthetic Electrocardiogram Signals
- McSharry, Clifford, et al.
- 2003
(Show Context)
Citation Context ...e sub-plots a to c correspond to the 5-minute time series for which a 5-second 4 The artificial RR interval generator detailed in this section has been previously published by the author in [189] and =-=[190]-=- 116sPower (dB) Power (dB) Power (dB) 0 −50 −100 −50 −100 a 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0 0 −50 −100 b 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 c 0 0.05 0.1 0.15 0.2 Frequency (Hz) 0.25 0.3 0.3... |

26 | Respiratory sinus arrhythmia in humans: how breathing pattern modulates heart rate - HIRSCH, BISHOP - 1981 |

23 |
FI, Mortara A, Schwartz PJ. Baroreflex sensitivity and heart-rate variability in prediction of total cardiac mortality after myocardial infarction
- MT, Jr, et al.
- 1998
(Show Context)
Citation Context ...owever, it is clear that the power in the HF band does carry information about the beat-to-beat variability of the heart and its ability to function in a healthy manner. For example, La Rovere et al. =-=[164]-=- and McClements [187] have shown that the HF as well as the LF metrics may be useful in long term prediction post-MI. Furthermore, the wealth of clinical studies that have presented evidence for the Ä... |

20 |
Auto-association by multi-layer perceptrons and singular value decomposition
- Boutlard, Kamp
- 1988
(Show Context)
Citation Context ...here �� � ÌË�Î Ø � �Ò� (3.28) Ï � Í�Ìs(3.29) with Ì being an arbitrary non-singular � ¢ � scaling matrix. Í� has � ¢ � elements, Ë� has � ¢ � elements and Î Ø has � ¢ Æ elements. It can be shown that =-=[37]-=- Ï � «sÌÍ Ø � (3.30) where Ï are the input-to-hidden layer weights and « is derived from a power series expansion of the activation function, � Ü � « « Ü for small Ü. For a linear activation function,... |

18 |
Magnitude and sign correlations in heartbeat fluctuations
- Ashkenazy, Ivanov, et al.
(Show Context)
Citation Context ...ve baseline shifts, but usually a scale must be chosen over which the detrending is performed and the HRV is over-estimated at this scale. Recently, scale-invariant techniques have also been proposed =-=[15, 16, 17, 29, 268]-=- to allow a measure of the fractal behaviour of the RR tachogram. These methods include: ¯ The Fano Factor [268]: the variance of the number of events in a specified counting time Ì divided by the mea... |

17 | Input Reconstruction Reliability Estimation - Pomerleau - 1993 |

16 |
Power spectral density of unevenly sampled data by least-square analysis: performance and application to heart rate signals
- Laguna, Moody, et al.
- 1998
(Show Context)
Citation Context ...al basis as in the conventional FFT. This technique is described in detail in the following section (Ü4.2). Several papers have presented results on HRV spectral estimation using the Lomb periodogram =-=[43, 162, 198]-=-. Laguna et al. [162] found that the Lomb method for estimating the PSD of unevenly sampled signals produces a better estimate of the PSD of an RR tachogram than re-sampling at 2Hz and using a convent... |

15 | Nikiforov IV. Detection of abrupt changes: Theory and Application - Basseville - 1993 |

15 |
Least squares, singular values and matrix approximations
- Golub
(Show Context)
Citation Context ...gonal matrix, Ë is an � ¢ Æ diagonal matrix with positive or zero elements (the singular values) and Î Ø is the transpose of an Æ ¢ Æ orthogonal matrix. The best rank-� approximation of � is given by =-=[93]-=- as Ï �� � Í�Ë�Î Ø � where �� � ÌË�Î Ø � �Ò� (3.28) Ï � Í�Ìs(3.29) with Ì being an arbitrary non-singular � ¢ � scaling matrix. Í� has � ¢ � elements, Ë� has � ¢ � elements and Î Ø has � ¢ Æ elements.... |

15 |
Courchesne E, Sejnowski TJ. Removal of eye activity artifacts from visual eventrelated potentials in normal and clinical subjects. Clin Neurophysiol 2000;111(10):1745–58
- TP, Makeig, et al.
(Show Context)
Citation Context ... analysis to be employed more universally. Recently, blind source separation, or independent component analysis (ICA), has been used to separate biomedical signals into their (independent) components =-=[22, 42, 135, 278, 283]-=-. ICA involves taking multiple observations of a signal which is a mixture of several statistically independent signals and identifying the independent signal sources (usually by minimising or maximis... |

15 | Nonlinear Noise Reduction for Electrocardiograms
- Schreiber, Kaplan
- 1996
(Show Context)
Citation Context ...involves a non-linear transformation which will distort the ECG waveform to some extent and in some scenarios (such as diagnosis), this may be highly significant. Although other non-linear techniques =-=[242, 250]-=-, have also been used to remove stationary non-correlated noise from ECGs, such methods have not been used to remove artefacts as yet. 3.3 Using morphological information to identify normal QRS comple... |

14 |
Correlated and uncorrelated regions in heart-rate fluctuations during sleep
- Bunde, Havlin, et al.
(Show Context)
Citation Context ...ose whose artery remains open. 41sMalik et al. [184] suggest that additional population based studies involving the full age spectrum for both sexes need to be performed. Furthermore, recent research =-=[38, 131]-=- has shown that guidelines need to be formulated which address the issues of how exercise, wakefulness and drug administration affect variability and how HRV can be meaningfully quantified in the ligh... |

14 | Software QRS detection in ambulatory monitoring-A review”, Med - Pahlm, Sornmo - 1984 |

13 |
Cardiovascular variability signals: towards the identification of a closed-loop model of the neural control mechanisms
- Baselli, Cerutti, et al.
- 1988
(Show Context)
Citation Context ... for the diastolic pressure to drop). This observation was first explained by DeBoer in 1987 [64] and although there have been other mathematical models proposed since then (such as the Baselli Model =-=[24]-=- and the Saul Model [238]), most experimental evidence is thought to support DeBoer’s model [255]. For a more detailed survey of the last 50 years of cardiovascular respiratory modelling the reader is... |

13 |
Vetterling W.T. and Flannery B.P. Numerical recipes in C. The art of scientific computing. Cambridge Univ
- Press, Teukolsky
(Show Context)
Citation Context ...ever, the meaning of the Nyquist frequency changes; the highest frequency about which there is information is � ¡ØÑ�Ò where ¡ØÑ�Ò � Ñ�Ò Ø�sØ� and a generalised Nyquist frequency is more commonly used =-=[229]-=-. This is calculated using ¡Ø�Ú, the average value of ¡Ø over all �’s and is therefore sometimes known as the average Nyquist frequency, � � � ¡Ø�Ú. Note that if the data is not zero-mean, unit varian... |

12 |
Establishing the relation between detrended fluctuation analysis and power spectral density analysis for stochastic processes, Phys
- Heneghan, McDarby
(Show Context)
Citation Context ... of the signal. In fact, in the case of stationary or quasi-stationary data, the above methods have been shown to give the same results as power spectral density methods. Furthermore, Heneghan et al. =-=[112]-=- have shown that there is a one-to-one correspondence between measures such as the DFA ÐÓ��� Ò ℄ slope ( ) and the PSD measured over the same time scales. The analysis in this thesis is therefore ÐÓ��... |

12 |
Applications of artificial neural networks for ECG signal detection and classification
- Hu, Tompkins, et al.
- 1993
(Show Context)
Citation Context ...f simple statistical methods to identify abnormal morphologies in the ECG leads to either a high number of false negatives or false positives. Preliminary studies by Jabri et al. [132], and Hu et al. =-=[125]-=- have shown that the application of ANNs to ECG analysis enables different QRS morphologies to be classified more accurately than with conventional algorithms. Hu et al. [125] investigated a 6-3-1 mul... |

12 | QRS morphology representation and noise estimation using the Karhunen–Loève transform - Moody, Mark - 1989 |

11 |
Transfer function analysis of the circulation: unique insights into cardiovascular regulation
- Saul, Berger, et al.
- 1991
(Show Context)
Citation Context ...ure to drop). This observation was first explained by DeBoer in 1987 [64] and although there have been other mathematical models proposed since then (such as the Baselli Model [24] and the Saul Model =-=[238]-=-), most experimental evidence is thought to support DeBoer’s model [255]. For a more detailed survey of the last 50 years of cardiovascular respiratory modelling the reader is referred to Batzel et al... |

10 | Why We Sleep: The Functions of Sleep in Humans and Other Mammals - Horne - 1988 |

10 | Testing for nonlinearity in unevenly sampled time series,” Phys - Schmitz, Schreiber - 1999 |

10 | Ventricular tachycardia and fibrillation detection by a sequential hypothesis testing algorithm - Thakor, YS, et al. - 1990 |

9 | Selective discrete Fourier transform algorithm for time-frequency analysis: method and application on simulated and cardiovascular signals - Keselbrener, Akselrod - 1996 |

9 | A.L.: Quantification of scaling exponents and crossover phenomena in nonstationary heartbeat time series - Peng, Havlin, et al. - 1995 |

8 |
Toward the early diagnosis of neonatal sepsis and sepsis-like illness using novel heart rate analysis
- Griffin, Moorman
- 2001
(Show Context)
Citation Context ...cent NN differing by more than � ms over an entire 24-hour ECG recording. 23sMany variations of these indices have been investigated with varying degrees of success. For instance, Griffin and Moorman =-=[98]-=- recently showed that neonatal sepsis is associated with a reduced baseline HR variability coupled with short-lived decelerations of HR and consequently the mean and SD of such distributions are often... |

8 |
Electronic evaluations of fetal heart rate patterns preceding fetal death, further observations
- Hon, Lee
- 1965
(Show Context)
Citation Context ...(HRV). Although the irregularity of the beat-to-beat intervals of a human heart was first noted in the early 1600’s [181], its physiological importance was not appreciated until 1965 when Hon and Lee =-=[119]-=- noted that fetal distress was preceded by changes in the pattern of beat-tobeat intervals before any significant change in the baseline heart rate. Since then many different indices and techniques ha... |

8 |
1994] “Comparison of methods for removal of ectopy in measurement of heart rate variability
- Lippman, Stein, et al.
(Show Context)
Citation Context ...al-to-Normal (N-N), intervals should be rejected from a study. However, he also acknowledges that this is an arbitrary threshold. Many different exclusion criteria have been proposed by other authors =-=[138, 147, 172, 181, 184]-=-, often with little justification. In many cases, where trials have been carried out to analyse the performance of algorithms using different RR interval exclusion criteria, the types of noise and art... |

7 |
Estimation of heart rate power spectrum bands from real-world data: dealing with ectopic beats and noisy data
- Albrecht, Cohen
(Show Context)
Citation Context ...r spectral components that can be distinguished more easily, and if the model order is chosen correctly can allow an accurate estimation of the PSD over very short windows. In 1989 Albrecht and Cohen =-=[10]-=- carried out experiments to compare the Fast Fourier Transforms (FFT) of linearly re-sampled RR tachograms with an autocorrelation method for constructing the PSD from only the original data points. T... |

7 | Sympathovagal balance: how should we measure it - Goldberger - 1999 |

7 |
A comparison of the predictive characteristics of heart rate variability index and left ventricular ejection fraction for all-cause mortality, arrhythmic events and sudden death after acute myocardial infarction
- Odemuyiwa, Malik, et al.
- 1991
(Show Context)
Citation Context ... depressed LVEF and increased ventricular ectopic activity) [184, 282]. ¯ Prediction of mortality - For prediction of all-cause mortality, HRV is similar to that of left ventricular ejection fraction =-=[184, 208]-=-. ¯ Prediction of arrhythmic events - HRV is superior to left ventricular ejection fraction in predicting arrhythmic events such as sudden cardiac death or ventricular tachycardia [184, 208] 22 partic... |

7 |
Dynamic heart rate variability: a tool for exploring sympathovagal balance continuously during sleep in men. Am J Physiol 1998;275(3 Pt 2):H946–50
- Otzenberger, Gronfier, et al.
(Show Context)
Citation Context ...cording to sleep states: awake (W), stage 2 (St2), slowwave sleep (SWS), and rapid eye movement sleep (REM). Ä� -ratio is also indicated for À� each physiological state. Taken from Otzenberger et al. =-=[213]-=-. ..............146 5.5 Two different expert sleep scores are plotted in blue and magenta for patient 02-01 from the OBS polysomnographic database. The horizontal time axis is in 30 second epochs. 1 r... |

7 | Detecting and analysing nonstationarity in a time series with non-linear cross-predictions, Phys - Schreiber - 1997 |

7 | Heart Rate Variability: Measures and Models
- Teich, Lowen, et al.
- 2001
(Show Context)
Citation Context ...ve baseline shifts, but usually a scale must be chosen over which the detrending is performed and the HRV is over-estimated at this scale. Recently, scale-invariant techniques have also been proposed =-=[15, 16, 17, 29, 268]-=- to allow a measure of the fractal behaviour of the RR tachogram. These methods include: ¯ The Fano Factor [268]: the variance of the number of events in a specified counting time Ì divided by the mea... |

6 |
Removing artifacts from electrocardiographic signals using independent components analysis
- Barros, Mansour, et al.
- 1998
(Show Context)
Citation Context ... analysis to be employed more universally. Recently, blind source separation, or independent component analysis (ICA), has been used to separate biomedical signals into their (independent) components =-=[22, 42, 135, 278, 283]-=-. ICA involves taking multiple observations of a signal which is a mixture of several statistically independent signals and identifying the independent signal sources (usually by minimising or maximis... |

6 |
Fusing conventional ECG QRS detection algorithms with an autoassociative neural network for the detection of ectopic beats
- GD, Tarassenko, et al.
- 2000
(Show Context)
Citation Context ... where the activation function is linear (�� � ) for the input and hidden layers; � � Æ� Ô� Â� �� �� �� � Ô � Û��sØ Ô � 7 Sections 3.3.5 to 3.3.11 have been previously published by Clifford et al. in =-=[54]-=- and [265] 93 (3.21)swhere Â is the number of hidden units. If this expression is differentiated with respect to Û�� and the derivative is set to zero the usual equations for least-squares optimisatio... |

6 |
RW: Basic Statistical Methods
- NM, Heath
- 1965
(Show Context)
Citation Context ...the mean Ä� -ratios for each of these sleep states is calculated for each À� spectral estimation method using the F-test, the Kolmogorov-Smirnov (K-S) test, and Student’s t-test for unequal variances =-=[229, 71]-=-. The F-test evaluates the probability that the variances in each distribution are significantly different (and hence one variance is significantly less than another). The K-S statistic is used for ca... |