28-06-2012, 12:54 PM
Nonlinear dynamics methods in the analysis of the heart rate variability
[attachment=25908]
Abstract
Purpose: We analyzed the heart rate variability (RR
intervals) by means of nonlinear dynamics methods: Poincaré
plot (return map), approximate entropy (ApEn) and
detrended fluctuation analysis (DFA). The purpose of this
study was the quantitative and qualitative assessment of
heart rate variability by means of these nonlinear dynamics
methods.
Material and methods: The Poincaré plot is a scattergram,
which is constructed by plotting each RR interval against
the previous one. Approximate entropy describes the complexity
and irregularity of the signals. Detrended fluctuation
analysis quantifies fractal-like correlation properties of the
data.
We analyzed two groups of patients: test group A – 15
diabetic children with diabetes type 1 and microalbuminuria
and control group C – 24 healthy children. For each patient
24 hour ECG (RR intervals) was recorded. Statistical analysis
was performed by means of nonparametric Mann-Whitney
test.
Results: Return maps of healthy children are mostly very
complex. In the case of diabetic children we found torpedoshaped
plots. The values of ApEn were lower in diabetic
children that indicated more regular heart rate in these
patients. DFA method shows also differences between the
investigated groups.
Conclusions: We concluded that using nonlinear dynamics
methods we could quantitatively and qualitatively study
the heart rate variability in healthy and diabetic patients.
INTRODUCTION
The Poincaré plot (return map) is a scattergram, which is
constructed by plotting each RR interval against the previous
one [4]. The Poincaré plot may be analyzed quantitatively by
fitting an ellipse to the plotted shape [5] (Fig. 1). The center of
the ellipse is determined by average RR interval. SD1 means
the standard deviation of the distances of points from y = x axis,
SD2 means the standard deviation of the distances of points
from y=-x+RR axis, where RR is the average R-R interval [6].
SD1 (instantaneous beat-to-beat variability of the data) determines
the width of the ellipse, SD2 (continuous beat-to-beat
variability) determines the length of the ellipse [7]. The ratio
SD1/SD2 is the measure of heart activity.
Approximate entropy (ApEn) describes the complexity and
irregularity of the signal [8,9]. ApEn is low in regular time series
and high in complex irregular ones. It can be applied to both
deterministic and stochastic signals and their combinations.
Detrended fluctuation analysis (DFA) quantifies fractal-like
correlation properties of the data [10]. The root-mean square
fluctuation of the integrated and detrended data are measured
in observation box of various sizes and then plotted against the
size of the box [11]. The scaling exponent represents the slope
of this line, which relates log(F(n)-fluctuation) to log(n-box
size). The short-term (F-fast) and long-term (S-slow) scaling
exponents are also calculated [12].
Material and methods
We analyzed two groups of patients: test group A – 15
diabetic children with diabetes type 1 and microalbuminuria
and control group C – 24 healthy children. For each patient
24 hour ECG was recorded. ECG records were divided into
two segments, day (06:00 to 22:00) and night activity (22:00 to
06:00), respectively (Anight, Aday, Cnight, Cday ).
Programs written in Matlab (MathWorks Inc., USA), a high
performance language for technical computing, were used to
analyze the ECG signals. The SD1/SD2 ratio, the approximate
entropy and DFA parameters were calculated. Statistical analysis
was performed by means of nonparametric Mann-Whitney
test for unpaired data.
Conclusions
We concluded that using nonlinear dynamics methods we
could quantitatively and qualitatively study the heart rate variability
in healthy and diabetic children.
[attachment=25908]
Abstract
Purpose: We analyzed the heart rate variability (RR
intervals) by means of nonlinear dynamics methods: Poincaré
plot (return map), approximate entropy (ApEn) and
detrended fluctuation analysis (DFA). The purpose of this
study was the quantitative and qualitative assessment of
heart rate variability by means of these nonlinear dynamics
methods.
Material and methods: The Poincaré plot is a scattergram,
which is constructed by plotting each RR interval against
the previous one. Approximate entropy describes the complexity
and irregularity of the signals. Detrended fluctuation
analysis quantifies fractal-like correlation properties of the
data.
We analyzed two groups of patients: test group A – 15
diabetic children with diabetes type 1 and microalbuminuria
and control group C – 24 healthy children. For each patient
24 hour ECG (RR intervals) was recorded. Statistical analysis
was performed by means of nonparametric Mann-Whitney
test.
Results: Return maps of healthy children are mostly very
complex. In the case of diabetic children we found torpedoshaped
plots. The values of ApEn were lower in diabetic
children that indicated more regular heart rate in these
patients. DFA method shows also differences between the
investigated groups.
Conclusions: We concluded that using nonlinear dynamics
methods we could quantitatively and qualitatively study
the heart rate variability in healthy and diabetic patients.
INTRODUCTION
The Poincaré plot (return map) is a scattergram, which is
constructed by plotting each RR interval against the previous
one [4]. The Poincaré plot may be analyzed quantitatively by
fitting an ellipse to the plotted shape [5] (Fig. 1). The center of
the ellipse is determined by average RR interval. SD1 means
the standard deviation of the distances of points from y = x axis,
SD2 means the standard deviation of the distances of points
from y=-x+RR axis, where RR is the average R-R interval [6].
SD1 (instantaneous beat-to-beat variability of the data) determines
the width of the ellipse, SD2 (continuous beat-to-beat
variability) determines the length of the ellipse [7]. The ratio
SD1/SD2 is the measure of heart activity.
Approximate entropy (ApEn) describes the complexity and
irregularity of the signal [8,9]. ApEn is low in regular time series
and high in complex irregular ones. It can be applied to both
deterministic and stochastic signals and their combinations.
Detrended fluctuation analysis (DFA) quantifies fractal-like
correlation properties of the data [10]. The root-mean square
fluctuation of the integrated and detrended data are measured
in observation box of various sizes and then plotted against the
size of the box [11]. The scaling exponent represents the slope
of this line, which relates log(F(n)-fluctuation) to log(n-box
size). The short-term (F-fast) and long-term (S-slow) scaling
exponents are also calculated [12].
Material and methods
We analyzed two groups of patients: test group A – 15
diabetic children with diabetes type 1 and microalbuminuria
and control group C – 24 healthy children. For each patient
24 hour ECG was recorded. ECG records were divided into
two segments, day (06:00 to 22:00) and night activity (22:00 to
06:00), respectively (Anight, Aday, Cnight, Cday ).
Programs written in Matlab (MathWorks Inc., USA), a high
performance language for technical computing, were used to
analyze the ECG signals. The SD1/SD2 ratio, the approximate
entropy and DFA parameters were calculated. Statistical analysis
was performed by means of nonparametric Mann-Whitney
test for unpaired data.
Conclusions
We concluded that using nonlinear dynamics methods we
could quantitatively and qualitatively study the heart rate variability
in healthy and diabetic children.