Clinical Validation of Machine Learning for Automatic Analysis of Multichannel Magnetocardiography

  • Riccardo Fenici
  • Donatella Brisinda
  • Anna Maria Meloni
  • Karsten Sternickel
  • Peter Fenici
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3504)


Magnetocardiographic (MCG) mapping measures magnetic fields generated by the electrophysiological activity of the heart. Quantitative analysis of MCG ventricular repolarization (VR) parameters may be useful to detect myocardial ischemia in patients with apparently normal ECG. However, manual calculation of MCG VR is time consuming and can be dependent on the examiner’s experience. Alternatively, the use of machine learning (ML) has been proposed recently to automate the interpretation of MCG recordings and to minimize human interference with the analysis. The aim of this study was to validate the predictive value of ML techniques in comparison with interactive, computer-aided, MCG analysis.

ML testing was done on a set of 140 randomly analysed MCG recordings from 74 subjects: 41 patients with ischemic heart disease (IHD) (group 1), 32 of them untreated (group 2), and 33 subjects without any evidence of cardiac disease (group 3). For each case at least 2 MCG datasets, recorded in different sessions, were analysed.

Two ML techniques combined identified abnormal VR in 25 IHD patients (group 1) and excluded VR abnormalities in 28 controls (group 3) providing 75% sensitivity, 85% specificity, 83% positive predictive value, 78% negative predictive value, 80% predictive accuracy This result was for the most part in agreement, but statistically better than that obtained with interactive analysis.

This study confirms that ML, applied on MCG recording at rest, has a predictive accuracy of 80% in detecting electrophysiological alterations associated with untreated IHD. Further work is needed to test the ML capability to differentiate VR alterations due to IHD from those due to non-ischemic cardiomyopathies.


Partial Little Square Predictive Accuracy Ventricular Repolarization Ischemic Heart Disease Patient Nuclear Stress Testing 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Tavarozzi, I., Comani, S., Del Gatta, C., Di Luzio, S., Romani, G.L., Gallina, S., et al.: Magnetocardiography: current status and perspectives. Part II: Clinical applications. Italian Heart J. 3(2), 151–165 (2002)Google Scholar
  2. 2.
    Fenici, R., Brisinda, D., Meloni, A.M., Fenici, P.: First 36-channel System for Clinical Magnetocardiography in Unshielded Hospital Laboratory for Cardiac Electrophysiology. International Journal of Bioelectromagnetism 5(1), 80–83 (2003)Google Scholar
  3. 3.
    Hänninen, H., Takala, P., Mäkijärvi, M., Montonen, J., Korhonen, P., Oikarinen, L., et al.: Detection of exercise induced myocardial ischemia by multichannel magnetocardiography in patients with single vessel coronary artery disease. A.N.E. 5(2), 147–157 (2000)Google Scholar
  4. 4.
    Tsukada, K., Miyashita, T., Kandori, A., Mitsui, T., Terada, Y., Sato, M., et al.: An iso-integral mapping technique using magnetocardiogram and its possible use for diagnosis of ischemic heart disease. Int. J. Card. Imaging 16, 55–66 (2000)CrossRefGoogle Scholar
  5. 5.
    Brisinda, D., Meloni, A.M., Fenici, P.: First 36-Channel Magnetocardiographic Study of CAD Patients in an Unshielded Laboratory for Interventional and Intensive Cardiac Care. In: Magnin, I.E., Montagnat, J., Clarysse, P., Nenonen, J., Katila, T. (eds.) FIMH 2003. LNCS, vol. 2674, pp. 122–131. Springer, Heidelberg (2003)CrossRefGoogle Scholar
  6. 6.
    Hailer, B., Chaicovsky, I., Auth-Eisernitz, S., Schafer, H., Steinberg, F., Grönemeyer, D.: Magnetocardiography in coronary artery disease with a new system in unshielded setting. Clin. Cardiol. 26(10), 465–471 (2003)CrossRefGoogle Scholar
  7. 7.
    Kanzaki, H., Nakatani, S., Kandori, A., Tsukada, K., Miyatake, K.: A new screening method to diagnose coronary artery disease using multichannel magnetocardiogram and simple exercise. Basic Res. Cardiol. 98(2), 124–132 (2003)CrossRefGoogle Scholar
  8. 8.
    Steinberg, B.A., Roguin, A., Allen, E., Wahl, D.R., Smith, C.R., St. John, M., et al.: Reproducibility and Interpretation of Magneto-Cardio-Gram Maps in Detecting Ischemia. The 53rd Annual Scientific Sessions of the American College of Cardiology, New Orleans, LA (March 2004); J. Am. Coll. Cardiol. Supplement 43, 149A (2004)Google Scholar
  9. 9.
    Tolstrup, K., Madsen, B., Brisinda, D., Meloni, A.M., Siegel, R., Smars, P.A., Fenici, R.: Resting Magnetocardiography Accurately Detects Myocardial Ischemia in Chest Pain Patients with normal or non-specific ECG Findings. Abstract N° 3440. Circulation Supplement 26 110(17), III–743 (2004)Google Scholar
  10. 10.
    Park, J.W., Reichert, U., Maleck, M., Klabes, J., Schafer, J., Jung, F.: Sensitivity and predictivity of magnetocardiography for the diagnosis of ischemic heart disease in patients with acute chest pain: preliminary results of Hoyerswerda Registry Study. Critical Pathways in Cardiology 1, 253–254 (2002)Google Scholar
  11. 11.
    Szymanski, B., Embrechts, M., Sternickel, K., Naenna, T., Bragaspathi, R.: Use of Machine Learning for Classification of Magnetocardiograms. In: Proceedings of the 2003 IEEE Conference on Systems, Man, and Cybernetics, SMC 2003, Washington, D. C., October 5-8, pp. 1400–1406 (2003)Google Scholar
  12. 12.
    CardioMag Imaging Inc. (CMI). Schenectady, USA. 36-channel system 2436 (Alfa version)Google Scholar
  13. 13.
    Sternickel, K., Tralshawala, N., Bakharev, A., et al.: Unshielded Measurements of Cardiac Electric Activity Using Magnetocardiography. International Journal of Bioelectromagnetism 4, 189–190 (2002)Google Scholar
  14. 14.
    Wold, S., Sjöström, M., Eriksson, L.: PLS-Regression: a Basic Tool of Chemometrics. Chemometrics and Intelligent Laboratory Systems 58, 109–130 (2001)CrossRefGoogle Scholar
  15. 15.
    Rosipal, R., Trejo, L.J.: Kernel Partial Least Squares Regression in Reproducing Kernel Hilbert Spaces. Journal of Machine Learning Research 2, 97–128 (2001)CrossRefGoogle Scholar
  16. 16.
    Embrechts, M., Szymanski, B., Sternickel, K.: A Brief Introduction to Scientific Data Mining: Direct Kernel Methods as a Fusion of Soft and Hard Computing. In: Ovasko, S. (ed.) Computationally Intelligent Hybrid Systems: The Fusion of Soft Computing and Hard Computing. IEEE Press, Los Alamitos (2004)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Riccardo Fenici
    • 1
  • Donatella Brisinda
    • 1
  • Anna Maria Meloni
    • 1
  • Karsten Sternickel
    • 1
  • Peter Fenici
    • 1
  1. 1.Clinical Physiology – Biomagnetism Research CenterCatholic University of Sacred HeartRomeItaly

Personalised recommendations