August 27, 2003Computers in cardiology
Multiscale entropy to distinguish physiologic and synthetic RR time series
Key result
A multiscale entropy method correctly classified 96% (48 of 50) of RR time series as physiologic or synthetic, and 100% when combined with Fourier spectral analysis.
Why the study?
Does a multiscale entropy method accurately distinguish physiologic from synthetic RR time series?
Population
RR time series datasets, including a learning set from healthy subjects and 50 test datasets from CinC 2002
Design
Other
Authors
Discussion
Loading...
Member takes
Implication
May aid RR data validation in research; leaves open prospective clinical validation.
Structured PICO
Does a multiscale entropy method accurately distinguish physiologic from synthetic RR time series?
P
Population
50 RR time series from the CinC 2002 test datasets used to validate a multiscale entropy method for distinguishing physiologic from synthetic signals.
I
Intervention
Multiscale entropy method (with or without Fourier spectral analysis)
O
Outcome
Correct classification of physiologic versus synthetic time seriessurrogate
Multiscale entropy, especially combined with Fourier spectral analysis, is highly effective at distinguishing physiologic from synthetic RR time series.
Cite This Study
Costa et al. (2003) studied RR time series classification (n=50). Multiscale entropy method was evaluated on Correct classification of physiologic vs synthetic time series. A multiscale entropy method correctly classified 96% (48 of 50) of RR time series as physiologic or synthetic, and 100% when combined with Fourier spectral analysis.