Medical & Biological Engineering & Computing

, Volume 53, Issue 3, pp 263–273 | Cite as

Application of the RIMARC algorithm to a large data set of action potentials and clinical parameters for risk prediction of atrial fibrillation

  • Ursula Ravens
  • Deniz Katircioglu-Öztürk
  • Erich Wettwer
  • Torsten Christ
  • Dobromir Dobrev
  • Niels Voigt
  • Claire Poulet
  • Simone Loose
  • Jana Simon
  • Agnes Stein
  • Klaus Matschke
  • Michael Knaut
  • Emre Oto
  • Ali Oto
  • H. Altay Güvenir
Original Article
  • 280 Downloads

Abstract

Ex vivo recorded action potentials (APs) in human right atrial tissue from patients in sinus rhythm (SR) or atrial fibrillation (AF) display a characteristic spike-and-dome or triangular shape, respectively, but variability is huge within each rhythm group. The aim of our study was to apply the machine-learning algorithm ranking instances by maximizing the area under the ROC curve (RIMARC) to a large data set of 480 APs combined with retrospectively collected general clinical parameters and to test whether the rules learned by the RIMARC algorithm can be used for accurately classifying the preoperative rhythm status. APs were included from 221 SR and 158 AF patients. During a learning phase, the RIMARC algorithm established a ranking order of 62 features by predictive value for SR or AF. The model was then challenged with an additional test set of features from 28 patients in whom rhythm status was blinded. The accuracy of the risk prediction for AF by the model was very good (0.93) when all features were used. Without the seven AP features, accuracy still reached 0.71. In conclusion, we have shown that training the machine-learning algorithm RIMARC with an experimental and clinical data set allows predicting a classification in a test data set with high accuracy. In a clinical setting, this approach may prove useful for finding hypothesis-generating associations between different parameters.

Keywords

Atrial fibrillation Risk prediction RIMARC algorithm Human right atrial action potentials Clinical parameters 

Abbreviations

AF

Atrial fibrillation

APA

Action potential amplitude (mV)

APD20

Action potential duration at 20 % of repolarization (ms)

APD50

Action potential duration at 50 % of repolarization (ms)

APD90

Action potential duration at 90 % of repolarization (ms)

dV/dtmax

Maximum rate of depolarization (Vs−1)

MAD

Maximum area under ROC curve-based discretization

PLT20

“Plateau potential” defined as the mean potential (mV) in the time window between 20 % of APD90 plus 5 ms

RIMARC

Ranking instances by maximizing the area under the ROC curve

RMP

Resting membrane potential (mV)

ROC

Receiver operating characteristics

SR

Sinus rhythm

Supplementary material

11517_2014_1232_MOESM1_ESM.doc (107 kb)
Supplementary material 1 (DOC 107 kb)
11517_2014_1232_MOESM2_ESM.doc (73 kb)
Supplementary material 2 (DOC 73 kb)
11517_2014_1232_MOESM3_ESM.xlsx (68 kb)
Supplementary material 3 (XLSX 68 kb)

References

  1. 1.
    Britton OJ, Bueno-Orovio A, Van AK, Lu HR, Towart R, Gallacher DJ et al (2013) Experimentally calibrated population of models predicts and explains intersubject variability in cardiac cellular electrophysiology. Proc Natl Acad Sci USA 110(23):E2098–E2105CrossRefPubMedCentralPubMedGoogle Scholar
  2. 2.
    Bustamante JO, Watanabe T, Murphy DA, McDonald TF (1982) Isolation of single atrial and ventricular cells from the human heart. Can Med Assoc J 126(7):791–793PubMedCentralPubMedGoogle Scholar
  3. 3.
    Dawodu AA, Monti F, Iwashiro K, Schiariti M, Chiavarelli R, Puddu PE (1996) The shape of human atrial action potential accounts for different frequency-related changes in vitro. Int J Cardiol 54(3):237–249CrossRefPubMedGoogle Scholar
  4. 4.
    Dobrev D, Ravens U (2003) Remodeling of cardiomyocyte ion channels in human atrial fibrillation. Basic Res Cardiol 98(3):137–148PubMedGoogle Scholar
  5. 5.
    Dobrev D, Wettwer E, Himmel HM, Kortner A, Kuhlisch E, Schuler S et al (2000) G-Protein beta(3)-subunit 825T allele is associated with enhanced human atrial inward rectifier potassium currents. Circulation 102(6):692–697CrossRefPubMedGoogle Scholar
  6. 6.
    Dorn A, Krueger MW, Seemann G, Doessel O (2012) Modelling of heterogeneous human atrial electrophysiology. Biomed Tech (Berl) 57(Suppl):1Google Scholar
  7. 7.
    Dossel O, Krueger MW, Weber FM, Wilhelms M, Seemann G (2012) Computational modeling of the human atrial anatomy and electrophysiology. Med Biol Eng Comput 50(8):773–799CrossRefPubMedGoogle Scholar
  8. 8.
    Eckstein J, Verheule S, de Groot N, Allessie M, Schotten U (2008) Mechanisms of perpetuation of atrial fibrillation in chronically dilated atria. Prog Biophys Mol Biol 97(2–3):435–451CrossRefPubMedGoogle Scholar
  9. 9.
    Gelband H, Bush HL, Rosen MR, Myerburg RJ, Hoffman BF (1972) Electrophysiologic properties of isolated preparations of human atrial myocardium. Circ Res 30(3):293–300CrossRefPubMedGoogle Scholar
  10. 10.
    Güvenir HA, Kurtcephe M (2013) Ranking instances by maximizing the area under ROC curve. IEEE Trans Knowl Data Eng 25(10):2356–2366CrossRefGoogle Scholar
  11. 11.
    Haissaguerre M, Jais P, Shah DC, Takahashi A, Hocini M, Quiniou G et al (1998) Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med 339(10):659–666CrossRefPubMedGoogle Scholar
  12. 12.
    Hill T, Lewicki P (2005) Statistics: methods and applications. A Comprehensive refeence for science, industry and data mining, 1st. 2005. StatSoft Inc., Tulsa, OKGoogle Scholar
  13. 13.
    Jakob H, Oelert H, Rupp J, Nawrath H (1989) Functional role of cholinoceptors and purinoceptors in human isolated atrial and ventricular heart muscle. Br J Pharmacol 97(4):1199–1208CrossRefPubMedCentralPubMedGoogle Scholar
  14. 14.
    Kirchhof P, Lip GY, Van Gelder IC, Bax J, Hylek E, Kaab S et al (2012) Comprehensive risk reduction in patients with atrial fibrillation: emerging diagnostic and therapeutic options–a report from the 3rd Atrial Fibrillation Competence NETwork/European Heart Rhythm Association consensus conference. Europace 14(1):8–27CrossRefPubMedCentralPubMedGoogle Scholar
  15. 15.
    Krueger MW, Dorn A, Keller DU, Holmqvist F, Carlson J, Platonov PG et al (2013) In-silico modeling of atrial repolarization in normal and atrial fibrillation remodeled state. Med Biol Eng Comput 51(10):1105–1119CrossRefPubMedGoogle Scholar
  16. 16.
    Kuijpers NH, Potse M, van Dam PM, ten Eikelder HM, Verheule S, Prinzen FW et al (2011) Mechanoelectrical coupling enhances initiation and affects perpetuation of atrial fibrillation during acute atrial dilation. Heart Rhythm 8(3):429–436CrossRefPubMedGoogle Scholar
  17. 17.
    Kurtcephe M, Güvenir HA (2013) A discretization method based on maximizing the area under receiver operating characteristic curve. Int J Pattern Recogn Artific Intel 27(Art. No. 1350002):1–26Google Scholar
  18. 18.
    Lam CS, Borlaug BA, Kane GC, Enders FT, Rodeheffer RJ, Redfield MM (2009) Age-associated increases in pulmonary artery systolic pressure in the general population. Circulation 119(20):2663–2670CrossRefPubMedCentralPubMedGoogle Scholar
  19. 19.
    Li GR, Nattel S (1997) Properties of human atrial ICa at physiological temperatures and relevance to action potential. Am J Physiol 272(1 Pt 2):H227–H235PubMedGoogle Scholar
  20. 20.
    Rottlaender D, Motloch LJ, Schmidt D, Reda S, Larbig R, Wolny M et al (2012) Clinical impact of atrial fibrillation in patients with pulmonary hypertension. PLoS ONE 7(3):e33902CrossRefPubMedCentralPubMedGoogle Scholar
  21. 21.
    Sanchez C, Bueno-Orovio A, Wettwer E, Loose S, Simon J, Ravens U et al (2014) Inter-subject variability in human atrial action potential in sinus rhythm versus chronic atrial fibrillation. PLoS ONE 9(8):e105897CrossRefPubMedCentralPubMedGoogle Scholar
  22. 22.
    Van Wagoner DR, Nerbonne JM (2000) Molecular basis of electrical remodeling in atrial fibrillation. J Mol Cell Cardiol 32(6):1101–1117CrossRefPubMedGoogle Scholar
  23. 23.
    Van Wagoner DR, Pond AL, Lamorgese M, Rossie SS, McCarthy PM, Nerbonne JM (1999) Atrial L-type Ca2+ currents and human atrial fibrillation. Circ Res 85(5):428–436CrossRefPubMedGoogle Scholar
  24. 24.
    Wettwer E, Hala O, Christ T, Heubach JF, Dobrev D, Knaut M et al (2004) Role of IKur in controlling action potential shape and contractility in the human atrium: influence of chronic atrial fibrillation. Circulation 110(16):2299–2306CrossRefPubMedGoogle Scholar
  25. 25.
    Workman AJ, Kane KA, Rankin AC (2008) Cellular bases for human atrial fibrillation. Heart Rhythm 5(6 Suppl):S1–S6CrossRefPubMedCentralPubMedGoogle Scholar

Copyright information

© International Federation for Medical and Biological Engineering 2014

Authors and Affiliations

  • Ursula Ravens
    • 1
    • 11
  • Deniz Katircioglu-Öztürk
    • 2
    • 3
  • Erich Wettwer
    • 1
  • Torsten Christ
    • 1
    • 4
  • Dobromir Dobrev
    • 1
    • 5
  • Niels Voigt
    • 1
    • 5
  • Claire Poulet
    • 1
    • 6
  • Simone Loose
    • 1
  • Jana Simon
    • 1
  • Agnes Stein
    • 7
  • Klaus Matschke
    • 8
  • Michael Knaut
    • 8
  • Emre Oto
    • 3
  • Ali Oto
    • 9
  • H. Altay Güvenir
    • 10
  1. 1.Department of Pharmacology and Toxicology, Medical Faculty Carl Gustav CarusTU DresdenDresdenGermany
  2. 2.Department of Medical Informatics, Informatics InstituteMiddle East Technical UniversityAnkaraTurkey
  3. 3.MITSAnkaraTurkey
  4. 4.Department of Experimental Pharmacology and ToxicologyUniversity Medical Center Hamburg-EppendorfHamburgGermany
  5. 5.Institute of Pharmacology, Faculty of MedicineUniversity of Duisburg-EssenEssenGermany
  6. 6.Imperial CollegeLondonUK
  7. 7.Department of AnesthesiologyHeart Center DresdenDresdenGermany
  8. 8.Clinic for Cardiac SurgeryHeart Center DresdenDresdenGermany
  9. 9.Department of CardiologyHacettepe University HospitalAnkaraTurkey
  10. 10.Department of Computer EngineeringBilkent UniversityAnkaraTurkey
  11. 11.Institut für Pharmakologie und ToxikologieTU DresdenDresdenGermany

Personalised recommendations