Advertisement

Machine learning approach in diagnosing Takotsubo cardiomyopathy: The role of the combined evaluation of atrial and ventricular strain, and parametric mapping

Published:November 17, 2022DOI:https://doi.org/10.1016/j.ijcard.2022.11.021

      Highlights

      • CMR-derived parameters were supposed to have high discriminatory power in identifying Takotsubo cardiomyopathy.
      • Many CMR-derived features were selected and implemented in a tree-based ML ensemble model to recognize TTC patients.
      • Left atrial strain parameters proved to be the best non-contrast CMR markers in making TTC diagnosis.

      Abstract

      Background

      Cardiac magnetic resonance (CMR) with late gadolinium enhancement (LGE) is a key diagnostic tool in the differential diagnosis between non-ischemic cause of cardiac chest pain. Some patients are not eligible for a gadolinium contrast-enhanced CMR; in this scenario, the diagnosis remains challenging without invasive examination. Our purpose was to derive a machine learning model integrating some non-contrast CMR parameters and demographic factors to identify Takotsubo cardiomyopathy (TTC) in subjects with cardiac chest pain.

      Material and methods

      Three groups of patients were retrospectively studied: TTC, acute myocarditis, and healthy controls. Global and regional left ventricular longitudinal, circumferential, and radial strain (RS) analysis included were assessed. Reservoir, conduit, and booster bi-atrial functions were evaluated by tissue-tracking. Parametric mapping values were also assessed in all the patients. Five different tree-based ensemble learning algorithms were tested concerning their ability in recognizing TTC in a fully cross-validated framework.

      Results

      The CMR-based machine learning (ML) ensemble model, by using the Extremely Randomized Trees algorithm with Elastic Net feature selection, showed a sensitivity of 92% (95% CI 78–100), specificity of 86% (95% CI 80–92) and area under the ROC of 0.94 (95% CI 0.90–0.99) in diagnosing TTC. Among non-contrast CMR parameters, the Shapley additive explanations analysis revealed that left atrial (LA) strain and strain rate were the top imaging markers in identifying TTC patients.

      Conclusions

      Our study demonstrated that using a tree-based ensemble learning algorithm on non-contrast CMR parameters and demographic factors enables the identification of subjects with TTC with good diagnostic accuracy.

      Translational outlook

      Our results suggest that non-contrast CMR features can be implemented in a ML model to accurately identify TTC subjects. This model could be a valuable tool for aiding in the diagnosis of subjects with a contraindication to the contrast media. Furthermore, the left atrial conduit strain and strain rate were imaging markers that had a strong impact on TTC identification. Further prospective and longitudinal studies are needed to validate these findings and assess predictive performance in different cohorts, such as those with different ethnicities, and social backgrounds and undergoing different treatments.

      Graphical abstract

      Keywords

      Abbreviations:

      AUROC (area under the receiver-operating characteristic curve), CMR (cardiac magnetic resonance), LA (left atrium), LV (left ventricle), LVEF (Left Ventricle Ejection Fraction), ML (machine learning), TTC (Takotsubo syndrome)
      To read this article in full you will need to make a payment

      Purchase one-time access:

      Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online access
      One-time access price info
      • For academic or personal research use, select 'Academic and Personal'
      • For corporate R&D use, select 'Corporate R&D Professionals'

      Subscribe:

      Subscribe to International Journal of Cardiology
      Already a print subscriber? Claim online access
      Already an online subscriber? Sign in
      Institutional Access: Sign in to ScienceDirect

      References

        • Ghadri J.-R.
        • Wittstein I.S.
        • Prasad A.
        • et al.
        International expert consensus document on Takotsubo syndrome (part I): clinical characteristics, diagnostic criteria, and pathophysiology.
        Eur. Heart J. 2018; 39: 2032-2046https://doi.org/10.1093/eurheartj/ehy076
        • Cau R.
        • Bassareo P.
        • Deidda M.
        • et al.
        Could CMR tissue-tracking and parametric mapping distinguish between takotsubo syndrome and acute myocarditis? A pilot study.
        Acad. Radiol. 2021; (Published online)
        • Lyon A.R.
        • Bossone E.
        • Schneider B.
        • et al.
        Current state of knowledge on Takotsubo syndrome: a position statement from the taskforce on Takotsubo syndrome of the heart failure association of the European Society of cardiology.
        Eur. J. Heart Fail. 2016; 18: 8-27https://doi.org/10.1002/ejhf.424
        • Eitel I.
        • von Knobelsdorff-Brenkenhoff F.
        • Bernhardt P.
        • et al.
        Clinical characteristics and cardiovascular magnetic resonance findings in stress (Takotsubo) cardiomyopathy.
        JAMA. 2011; 306https://doi.org/10.1001/jama.2011.992
        • Collet J.-P.
        • Thiele H.
        • Barbato E.
        • et al.
        2020 ESC guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: the task force for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation.
        Eur. Heart J. 2021; 42: 1289-1367https://doi.org/10.1093/eurheartj/ehaa575
        • Eitel I.
        • Lücke C.
        • Grothoff M.
        • et al.
        Inflammation in takotsubo cardiomyopathy: insights from cardiovascular magnetic resonance imaging.
        Eur. Radiol. 2010; 20: 422-431https://doi.org/10.1007/s00330-009-1549-5
        • Cau R.
        • Solinas C.
        • De Silva P.
        • et al.
        Role of cardiac MRI in the diagnosis of immune checkpoint inhibitor-associated myocarditis.
        Int. J. Cancer. June 2022; (Published online)https://doi.org/10.1002/ijc.34169
        • Cau R.
        • Bassareo P.
        • Caredda G.
        • Suri J.S.
        • Esposito A.
        • Saba L.
        Atrial strain by feature-tracking cardiac magnetic resonance imaging in takotsubo cardiomyopathy. features, feasibility, and reproducibility.
        Can. Assoc. Radiol. J. October 2021; (Published online. 8465371211042497)https://doi.org/10.1177/08465371211042497
        • Scatteia A.
        • Baritussio A.
        • Bucciarelli-Ducci C.
        Strain imaging using cardiac magnetic resonance.
        Heart Fail. Rev. 2017; 22: 465-476https://doi.org/10.1007/s10741-017-9621-8
        • Palmisano V.
        • Cossa S.
        • Esposito A.
        • et al.
        Obstructive and nonobstructive hypertrophic cardiomyopathy: differences in global and segmental myocardial strain by cardiac magnetic resonance feature tracking.
        J. Thorac. Imaging. 2022; 37: 49-57https://doi.org/10.1097/RTI.0000000000000612
        • Cau R.
        • Loewe C.
        • Cherchi V.
        • et al.
        Atrial impairment as a marker in discriminating between takotsubo and acute myocarditis using cardiac magnetic resonance.
        J. Thorac. Imaging. 2022; (Publish Ah(00):1–7)https://doi.org/10.1097/rti.0000000000000650
        • Cau R.
        • Bassareo P.
        • Suri J.S.
        • Pontone G.
        • Saba L.
        • Saba L.
        The emerging role of atrial strain assessed by cardiac MRI in different cardiovascular settings: an up-to-date review.
        Feature Track. 2022; 1
        • Xu J.
        • Yang W.
        • Zhao S.
        • Lu M.
        State-of-the-art myocardial strain by CMR feature tracking: clinical applications and future perspectives.
        Eur. Radiol. 2022; 32: 5424-5435https://doi.org/10.1007/s00330-022-08629-2
        • Kato K.
        • Daimon M.
        • Sano M.
        • et al.
        Dynamic trend of myocardial edema in takotsubo syndrome: a serial cardiac magnetic resonance study.
        J. Clin. Med. 2022; 11https://doi.org/10.3390/jcm11040987
        • Cau R.
        • Cherchi V.
        • Micheletti G.
        • et al.
        Potential role of artificial intelligence in cardiac magnetic resonance imaging.
        J. Thorac. Imaging. 2021; (Publish Ah(3):142–148)https://doi.org/10.1097/rti.0000000000000584
        • Cau R.
        • Flanders A.
        • Mannelli L.
        • et al.
        Artificial intelligence in computed tomography plaque characterization: a review.
        Eur. J. Radiol. 2021; (Published online): 109767https://doi.org/10.1016/j.ejrad.2021.109767
        • Muscogiuri G.
        • Volpato V.
        • Cau R.
        • et al.
        Application of AI in cardiovascular multimodality imaging.
        Heliyon. 2022; (Published online): e10872https://doi.org/10.1016/j.heliyon.2022.e10872
        • Caforio A.L.P.
        • Pankuweit S.
        • Arbustini E.
        • et al.
        Current state of knowledge on aetiology, diagnosis, management, and therapy of myocarditis: a position statement of the European Society of Cardiology Working Group on myocardial and pericardial diseases.
        Eur. Heart J. 2013; 34 (2648a-2648d): 2636-2648
        • Breiman L.
        Random forests.
        Mach. Learn. 2001; 45: 5-32https://doi.org/10.1023/A:1010933404324
        • Geurts P.
        • Ernst D.
        • Wehenkel L.
        Extremely randomized trees.
        Mach. Learn. 2006; 63: 3-42https://doi.org/10.1007/s10994-006-6226-1
        • Breiman L.
        Bagging predictors.
        Mach. Learn. 1996; 24: 123-140https://doi.org/10.1007/BF00058655
        • Opitz D.W.
        • Shavlik J.W.
        Actively searching for an effective neural network ensemble.
        Connect. Sci. 1996; 8: 337-354https://doi.org/10.1080/095400996116802
        • Biglari M.
        • Mirzaei F.
        • Hassanpour H.
        Feature selection for small sample sets with high dimensional data using heuristic hybrid approach.
        Int. J. Eng. 2020; 33: 213-220https://doi.org/10.5829/ije.2020.33.02b.05
        • Zou H.
        • Hastie T.
        Regularization and variable selection via the elastic net.
        J. R. Stat. Soc. 2005; 67: 301-320
        • Brown P.J.
        • Zhong J.
        • Frood R.
        • et al.
        Prediction of outcome in anal squamous cell carcinoma using radiomic feature analysis of pre-treatment FDG PET-CT.
        Eur. J. Nucl. Med. Mol. Imaging. 2019; 46: 2790-2799https://doi.org/10.1007/s00259-019-04495-1
        • Lundberg S.M.
        • Lee S.-I.
        A unified approach to interpreting model predictions.
        in: Proceedings of the 31st International Conference on Neural Information Processing Systems. NIPS’17. Curran Associates Inc, 2017: 4768-4777
        • DeLong E.R.
        • DeLong D.M.
        • Clarke-Pearson D.L.
        Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach.
        Biometrics. 1988; 44: 837-845
        • Assel M.
        • Sjoberg D.D.
        • Vickers A.J.
        The brier score does not evaluate the clinical utility of diagnostic tests or prediction models.
        Diagnos. Progn Res. 2017; 1: 19https://doi.org/10.1186/s41512-017-0020-3
        • Lossnitzer D.
        • Klenantz S.
        • Andre F.
        • et al.
        Stable patients with suspected myocardial ischemia: comparison of machine-learning computed tomography-based fractional flow reserve and stress perfusion cardiovascular magnetic resonance imaging to detect myocardial ischemia.
        BMC Cardiovasc. Disord. 2022; 22: 34https://doi.org/10.1186/s12872-022-02467-2
        • Sundaram D.S.B.
        • Arunachalam S.P.
        • Damani D.N.
        • et al.
        Natural Language Processing Based Machine Learning Model Using Cardiac MRI Reports to Identify Hypertrophic Cardiomyopathy Patients.
        (Published online)April 12, 2021https://doi.org/10.1115/DMD2021-1076
        • Overmars L.M.
        • van Es B.
        • Groepenhoff F.
        • et al.
        Preventing unnecessary imaging in patients suspect of coronary artery disease through machine learning of electronic health records.
        Eur Hear J - Digit Heal. 2022; 3: 11-19https://doi.org/10.1093/ehjdh/ztab103
        • Baeßler B.
        • Mannil M.
        • Maintz D.
        • Alkadhi H.
        • Manka R.
        Texture analysis and machine learning of non-contrast T1-weighted MR images in patients with hypertrophic cardiomyopathy-preliminary results.
        Eur. J. Radiol. 2018; 102: 61-67https://doi.org/10.1016/j.ejrad.2018.03.013
        • Gopalakrishnan V.
        • Menon P.G.
        • Madan S.
        cMRI-BED: A novel informatics framework for cardiac MRI biomarker extraction and discovery applied to pediatric cardiomyopathy classification.
        Biomed. Eng. Online. 2015; 14 Suppl 2: S7https://doi.org/10.1186/1475-925X-14-S2-S7
        • Hinojar R.
        • Zamorano J.L.
        • Fernández-Méndez M.A.
        • et al.
        Prognostic value of left atrial function by cardiovascular magnetic resonance feature tracking in hypertrophic cardiomyopathy.
        Int. J. Card. Imaging. 2019; 35: 1055-1065https://doi.org/10.1007/s10554-019-01534-8
        • Dick A.
        • Schmidt B.
        • Michels G.
        • Bunck A.C.
        • Maintz D.
        • Baeßler B.
        Left and right atrial feature tracking in acute myocarditis: a feasibility study.
        Eur. J. Radiol. 2017; 89: 72-80https://doi.org/10.1016/j.ejrad.2017.01.028
        • Doerner J.
        • Bunck A.C.
        • Michels G.
        • Maintz D.
        • Baeßler B.
        Incremental value of cardiovascular magnetic resonance feature tracking derived atrial and ventricular strain parameters in a comprehensive approach for the diagnosis of acute myocarditis.
        Eur. J. Radiol. 2018; 104: 120-128https://doi.org/10.1016/j.ejrad.2018.05.012
        • Backhaus S.J.
        • Stiermaier T.
        • Lange T.
        • et al.
        Atrial mechanics and their prognostic impact in Takotsubo syndrome: a cardiovascular magnetic resonance imaging study.
        Eur. Heart J. Cardiovasc. Imaging. 2019; 20: 1059-1069https://doi.org/10.1093/ehjci/jey219
        • Stiermaier T.
        • Graf T.
        • Möller C.
        • et al.
        Transient left atrial dysfunction is a feature of Takotsubo syndrome.
        J. Cardiovasc. Magn. Reson. 2017; 19: 15https://doi.org/10.1186/s12968-017-0328-8
        • Stiermaier T.
        • Lange T.
        • Chiribiri A.
        • et al.
        Left ventricular myocardial deformation in Takotsubo syndrome: a cardiovascular magnetic resonance myocardial feature tracking study.
        Eur. Radiol. 2018; 28: 5160-5170https://doi.org/10.1007/s00330-018-5475-2
        • von Roeder M.
        • Rommel K.-P.
        • Kowallick J.T.
        • et al.
        Influence of left atrial function on exercise capacity and left ventricular function in patients with heart failure and preserved ejection fraction.
        Circ. Cardiovasc. Imag. 2017; 10e005467https://doi.org/10.1161/CIRCIMAGING.116.005467
        • Emerson R.W.
        • Adams C.
        • Nishino T.
        • et al.
        Functional neuroimaging of high-risk 6-month-old infants predicts a diagnosis of autism at 24 months of age.
        Sci. Transl. Med. 2017; 9https://doi.org/10.1126/scitranslmed.aag2882