Abstract
The aim of this study was to analyze the whole temporal profiles of the segmental deformation curves of the left ventricle (LV) and describe their interrelations to obtain more detailed information concerning global LV function in order to be able to identify abnormal changes in LV mechanics. The temporal characteristics of the segmental LV deformation curves were compactly described using an efficient decomposition into major patterns of variation through a statistical method, called Principal Component Analysis (PCA). In order to describe the spatial relations between the segmental traces, the PCA-derived temporal features of all LV segments were concatenated. The obtained set of features was then used to build an automatic classification system. The proposed methodology was applied to a group of 60 MRI-delayed enhancement confirmed infarct patients and 60 controls in order to detect myocardial infarction. An average classification accuracy of 87% with corresponding sensitivity and specificity rates of 89% and 85%, respectively was obtained by the proposed methodology applied on the strain rate curves. This classification performance was better than that obtained with the same methodology applied on the strain curves, reading of two expert cardiologists as well as comparative classification systems using only the spatial distribution of the end-systolic strain and peak-systolic strain rate values. This study shows the potential of machine learning in the field of cardiac deformation imaging where an efficient representation of the spatio-temporal characteristics of the segmental deformation curves allowed automatic classification of infarcted from control hearts with high accuracy.
Similar content being viewed by others
References
Kvitting J-PE, Wigstroom L, Strotmann JM, Sutherland GR (1999) How accurate is visual assessment of synchronicity in myocardial motion? An in vitro study with computer-simulated regional delay in myocardial motion: clinical implications for rest and stress echocardiography studies. J Am Soc Echocardiogr 12:698–705
Hoffmann R, Lethen H, Marwick T et al (1996) Analysis of interinstitutional observer agreement in interpretation of dobutamine stress echocardiograms. J Am Coll Cardiol 27:330–333
Sutherland GR, Di Salvo G, Claus P, D’hooge J, Bijnens B (2004) Strain and strain rate imaging: a new clinical approach to quantifying regional myocardial function. J Am Soc Echocardiogr 17:788–802
Delgado V, Ypenburg C, van Bommel RJ et al (2008) Assessment of left ventricular dyssynchrony by speckle tracking strain imaging: comparison between longitudinal, circumferential, and radial strain in cardiac resynchronization therapy. J Am Coll Cardiol 51:1944–1952
Bax JJ, Molhoek SG, Van Erven L et al (2003) Usefulness of myocardial tissue Doppler echocardiography to evaluate left ventricular dyssynchrony before and after biventricular pacing in patients with idiopathic dilated cardiomyopath. Am J Cardiol 91:94–97
Tabassian M, Alessandrini M, Herbots L, et al. (2015) Automatic detection of ischemic myocardium by spatio-temporal analysis of echocardiographic strain and strain rate curves. IEEE Int Ultrason Symp 1–4
Rademakers F, Engvall J, Edvardsen T et al (2013) Determining optimal noninvasive parameters for the prediction of left ventricular remodeling in chronic ischemic patients. Scand Card J 47:329–334
Herbots L (2006) Quantification of regional myocardial deformation, normal characteristics and clinical use in ischaemic heart disease. Leuven University Press, Leuven (ISBN: 9789058675552)
Janssens S, Dubois C, Bogaert J et al (2006) Autologous bone marrow-derived stem-cell transfer in patients with ST-segment elevation myocardial infarction: double-blind, randomized controlled trial. Lancet 367:113–121
Desmet W, Bogaert J, Dubois C et al (2011) High-dose intracoronary adenosine for myocardial salvage in patients with acute ST-segment elevation myocardial infarction. Eur Heart J 32:867–877
Wright J, Adriaenssens T, Dymarkowski S, Desmet W, Bogaert J (2009) Quantification of myocardial area at risk with T2-weighted CMR, comparison with contrast-enhanced CMR and coronary angiography. JACC 7:825–831
Lang RM, Badano, LP, Mor-Avi V et al (2015) Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. Eur Heart J 16: 233–271
Claus P, D’hooge J, Langeland T, Bijnens B, Sutherland G (2002) SPEQLE (Software Package for Echocardiographic Quantification LEuven) an integrated approach to ultrasound-based cardiac deformation quantification. IEEE Comput Cardiol 69–72. http://ieeexplore.ieee.org/document/1166709/
D’hooge J, Bijnens B, Thoen J, Van de Werf F, Sutherland GR, Suetens P (2002) Echocardiographic strain and strain-rate imaging: a new tool to study regional myocardial function. IEEE Trans Med Imaging 21:1022–1030
Jolliffe I (2002) Principal component analysis. Wiley, Hoboken
Tabassian M, Alessandrini M, De Marchi L et al (2015) Principal component analysis for the classification of cardiac motion abnormalities based on echocardiographic strain and strain rate imaging, Vol 9126. The 8th International Conference on Functional Imaging and Modeling of the Heart. Springer, New York, pp 83–90
Cover T, Hart P (1967) Nearest neighbor pattern classification. IEEE Trans Inf Theory 13:21–27
Lalkhen AG, McCluskey A (2008) Clinical tests: sensitivity and specificity. Continuing education in anaesthesia. Crit Care Pain 8:221–223
Dietterich TG (1998) Approximate statistical tests for comparing supervised classification learning algorithms. Neural Comput 10:1895–1923
Ingul CB, Stoylen A, Slordahl SA, Wiseth R, Burgess M, Marwick TH (2007) Automated analysis of myocardial deformation at dobutamine stress echocardiography. J Am Coll Cardiol 49:1651–1659
Ferferieva V, Van den Bergh A, Claus P et al (2012) The relative value of strain and strain rate for defining intrinsic myocardial function. Am J Physiol Heart Circ Physiol 302:188–195
Abraham TP, Dimaano VL, Liang H (2007) Role of tissue Doppler and strain echocardiography in current clinical practice. Circulation 116:2597–2609
Acknowledgements
The authors would like to thank Prof. Stefan Janssens, principal investigator of the stem-cell study, and Prof. Walter Desmet, principal investigator of the SALVAGE study, (both from the Department of Cardiovascular Sciences, KU Leuven, Belgium) for providing us with their databases.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
None.
Rights and permissions
About this article
Cite this article
Tabassian, M., Alessandrini, M., Herbots, L. et al. Machine learning of the spatio-temporal characteristics of echocardiographic deformation curves for infarct classification. Int J Cardiovasc Imaging 33, 1159–1167 (2017). https://doi.org/10.1007/s10554-017-1108-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10554-017-1108-0