Medical and Biological Engineering and Computing

, Volume 39, Issue 6, pp 664–671 | Cite as

Automatic detection of atrial fibrillation using the coefficient of variation and density histograms of RR and ΔRR intervals



The paper describes a method for the automatic detection of atrial fibrillation, an abnormal heart rhythm, based on the sequence of intervals between heartbeats. The RR interval is the interbeat interval, and ΔRR is the difference between two successive RR intervals. Standard density histograms of the RR and ΔRR intervals were prepared as templates for atrial fibrillation detection. As the coefficients of variation of the RR and ΔRR intervals were approximately constant during atrial fibrillation, the coefficients of variation in the test data could be compared with the standard coefficients of variation (CV test). Further, the similarities between the density histograms of the test data and the standard density histograms were estimated using the Kolmogorov-Smirnov test. The CV test based on the RR intervals showed a sensitivity of 86.6% and a specificity of 84.3%. The CV test based on the ΔRR intervals showed that the sensitivity and the specificity are both approximately 84%. The Kolmogorov-Smirnov test based on the RR intervals did not improve on the result of the CV test. In contrast, the Kolmogorov-Smirnov test based on the ΔRR intervals showed a sensitivity of 94.4% and a specificity of 97.2%.


Atrial fibrillation RR interval Coefficient of variation Kolmogorov-Smirnov test 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Andresen, D., andBrüggemann, T. (1998): ‘Heat rate variability preceding onset of atrial fibrillation’,J. Cardiovasc. Electrophysiol. Supp.,9, pp. S26-S29Google Scholar
  2. Bootsma, B., Hoolen, A., Strackee, J., andMeijler, F. (1970): ‘Analysis of R-R intervals in patients with atrial fibrillation at rest and during exercise,’Circulation,41, pp. 783–794Google Scholar
  3. Hulley, S., andCumming, S. (Eds) (1988): ‘Designing clinical research’ (Williams & Wilkins, 1988)Google Scholar
  4. Meijler, F., Jalife, J., Beaumont, J., andVaidya, D. (1996): ‘AV nodal function during atrial fibrillation: the role of electronic modulation of propagation,’J. Cardiovasc. Electrophysiol.,7, pp. 843–861Google Scholar
  5. Meijler, F., andWittkampf, F. (1997): ‘Role of the atrioventricular node in atrial fibrillation’ inFalk, R., andPodrid, P. (Eds): ‘Atrial fibrillation: mechanisms and management, 2nd edn’ (Loppincott-Raven Publishers, Philadelphia, 1997), pp. 109–131Google Scholar
  6. Moody, G., andMark, R. (1983): ‘A new method for detecting atrial fibrillation using R-R intervals’,Comput. Cardiol. pp. 227–230Google Scholar
  7. Murgatroyd, F., Xie, B., Copie, X., Blankoff, I., Camm, A., andMalik, M. (1995): ‘Identification of atrial fibrillation episodes in ambulatory electrocardiographic recordings: validation of a method for obtaining labeled R-R interval files’,Pacing Clin. Electrophysiol.,18, pp. 1315–1320Google Scholar
  8. Pinciroli, F., andCastelli, A. (1986): ‘Pre-clinical experimentation of a quantitative synthesis of the local variability in the original R-R interval sequence in the presence of arrhythmia’,Automedica,6, pp. 295–317Google Scholar
  9. Press, W., Teukolsky, S., Vetterling, W., andFlannery, B. (Eds) (1992): ‘Numerical recipes in C: The art of scientific computing’ (Cambridge University Press, 1992)Google Scholar
  10. Slocum, J., Sahakian, A., andSwiryn, S. (1987): ‘Computer detection of atrial fibrillation on the surface electrocardiogram’,Comput. Cardiol.,13, pp. 253–254Google Scholar
  11. Tateno, K., andGlass, L. (2000): ‘A method for detection of atrial fibrillation using RR intervals’,Comput. Cardiol.,27, pp. 391–394Google Scholar

Copyright information

© IFMBE 2001

Authors and Affiliations

  1. 1.Department of PhysiologyMcGill UniversityMontrealCanada

Personalised recommendations