CinE caRdiac magneTic resonAnce to predIct veNTricular arrhYthmia (CERTAINTY)

Julian Krebs, Tommaso Mansi, Hervé Delingette, Bin Lou, Joao A C Lima, Susumu Tao, Luisa A Ciuffo, Sanaz Norgard, Barbara Butcher, Wei H Lee, Ela Chamera, Timm-Michael Dickfeld, Michael Stillabower, Joseph E Marine, Robert G Weiss, Gordon F Tomaselli, Henry Halperin, Katherine C Wu, Hiroshi Ashikaga, Julian Krebs, Tommaso Mansi, Hervé Delingette, Bin Lou, Joao A C Lima, Susumu Tao, Luisa A Ciuffo, Sanaz Norgard, Barbara Butcher, Wei H Lee, Ela Chamera, Timm-Michael Dickfeld, Michael Stillabower, Joseph E Marine, Robert G Weiss, Gordon F Tomaselli, Henry Halperin, Katherine C Wu, Hiroshi Ashikaga

Abstract

Better models to identify individuals at low risk of ventricular arrhythmia (VA) are needed for implantable cardioverter-defibrillator (ICD) candidates to mitigate the risk of ICD-related complications. We designed the CERTAINTY study (CinE caRdiac magneTic resonAnce to predIct veNTricular arrhYthmia) with deep learning for VA risk prediction from cine cardiac magnetic resonance (CMR). Using a training cohort of primary prevention ICD recipients (n = 350, 97 women, median age 59 years, 178 ischemic cardiomyopathy) who underwent CMR immediately prior to ICD implantation, we developed two neural networks: Cine Fingerprint Extractor and Risk Predictor. The former extracts cardiac structure and function features from cine CMR in a form of cine fingerprint in a fully unsupervised fashion, and the latter takes in the cine fingerprint and outputs disease outcomes as a cine risk score. Patients with VA (n = 96) had a significantly higher cine risk score than those without VA. Multivariate analysis showed that the cine risk score was significantly associated with VA after adjusting for clinical characteristics, cardiac structure and function including CMR-derived scar extent. These findings indicate that non-contrast, cine CMR inherently contains features to improve VA risk prediction in primary prevention ICD candidates. We solicit participation from multiple centers for external validation.

Conflict of interest statement

HA serves as a consultant to Siemens Healthineers. No other authors have competing interests.

© 2021. The Author(s).

Figures

Figure 1
Figure 1
Algorithm overview. (A) Cine fingerprint extractor. (B) Risk predictor. See text for details.
Figure 2
Figure 2
Survival prediction for each endpoint. (A) LV LGE gray zone. (B) LA maximum volume index, (C) LA total emptying fraction, (D) Cine risk score. The shaded area represents 95% confidence interval.
Figure 3
Figure 3
Survival prediction for each endpoint using multivariate Cox Proportional Hazards Regression models. (A) Cine risk score + LV LGE gray zone. (B) Cine risk score + LA maximum volume index. (C) Cine risk score + LA total emptying fraction. (D) LV LGE gray zone + LA maximum volume index + LA total emptying fraction. (E) Cine risk score + LV LGE gray zone + LA maximum volume index + LA total emptying fraction. The shaded area represents 95% confidence interval.

References

    1. Epstein AE, et al. 2012 ACCF/AHA/HRS focused update incorporated into the ACCF/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities: A report of the American college of cardiology foundation/american heart association task force on practice guide. Circulation. 2013 doi: 10.1161/CIR.0b013e318276ce9b.
    1. Sabbag A, et al. Contemporary rates of appropriate shock therapy in patients who receive implantable device therapy in a real-world setting: From the Israeli ICD Registry. Heart Rhythm. 2015;12:2426–2433. doi: 10.1016/j.hrthm.2015.08.020.
    1. Prutkin JM, et al. Rates of and factors associated with infection in 200 909 medicare implantable cardioverter-defibrillator implants results from the national cardiovascular data registry. Circulation. 2014 doi: 10.1161/CIRCULATIONAHA.114.009081.
    1. Moss AJ, et al. Reduction in inappropriate therapy and mortality through ICD programming. N. Engl. J. Med. 2012 doi: 10.1056/NEJMoa1211107.
    1. Poole JE, et al. Prognostic importance of defibrillator shocks in patients with heart failure. N. Engl. J. Med. 2008 doi: 10.1056/NEJMoa071098.
    1. Schron EB, et al. Quality of life in the antiarrhythmics versus implantable defibrillators trial: Impact of therapy and influence of adverse symptoms and defibrillator shocks. Circulation. 2002 doi: 10.1161/hc0502.103330.
    1. Adabag S, et al. Association of implantable cardioverter defibrillators with survival in patients with and without improved ejection fraction. JAMA Cardiol. 2017 doi: 10.1001/jamacardio.2017.1413.
    1. Wu KC, et al. Combined cardiac magnetic resonance imaging and C-reactive protein levels identify a cohort at low risk for defibrillator firings and death. Circ. Cardiovasc. Imaging. 2012;5:178–186. doi: 10.1161/CIRCIMAGING.111.968024.
    1. Mikami Y, et al. Right ventricular ejection fraction is incremental to left ventricular ejection fraction for the prediction of future arrhythmic events in patients with systolic dysfunction. Circ. Arrhythmia Electrophysiol. 2017 doi: 10.1161/CIRCEP.116.004067.
    1. Lee D, Parkash R. Left atrial function and sudden cardiac death. Can. J. Cardiol. 2019;35:1091–1093. doi: 10.1016/j.cjca.2019.05.022.
    1. Ardila D, et al. End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography. Nat. Med. 2019 doi: 10.1038/s41591-019-0447-x.
    1. Lou B, et al. An image-based deep learning framework for individualising radiotherapy dose: A retrospective analysis of outcome prediction. Lancet Digit. Health. 2019;1:e136–e147. doi: 10.1016/S2589-7500(19)30058-5.
    1. Bello GA, et al. Deep-learning cardiac motion analysis for human survival prediction. Nat. Mach. Intell. 2019;1:95–104. doi: 10.1038/s42256-019-0019-2.
    1. Krebs J, Mansi T, Ayache N, Delingette H. Probabilistic motion modeling from medical image sequences: Application to cardiac cine-MRI. Lect. Notes Comput. Sci. 2020 doi: 10.1007/978-3-030-39074-7_19.
    1. Kingma, D. P., Rezende, D. J., Mohamed, S. & Welling, M. Semi-supervised learning with deep generative models. in Advances in Neural Information Processing Systems (2014).
    1. Wu KC. Sudden cardiac death substrate imaged by magnetic resonance imaging: From investigational tool to clinical applications. Circ. Cardiovasc. Imaging. 2017 doi: 10.1161/CIRCIMAGING.116.005461.
    1. Ashikaga H, et al. Feasibility of image-based simulation to estimate ablation target in human ventricular arrhythmia. Heart Rhythm. 2013;10:1109–1116. doi: 10.1016/j.hrthm.2013.04.015.
    1. Rijnierse MT, et al. Usefulness of left atrial emptying fraction to predict ventricular arrhythmias in patients with implantable cardioverter defibrillators. Am. J. Cardiol. 2017;120:243–250. doi: 10.1016/j.amjcard.2017.04.015.
    1. Negishi K, et al. Left atrial booster pump function is an independent predictor of subsequent life-threatening ventricular arrhythmias in non-ischaemic cardiomyopathy. Eur. Heart J. Cardiovasc. Imaging. 2016 doi: 10.1093/ehjci/jev333.
    1. Gulshan V, et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA J. Am. Med. Assoc. 2016 doi: 10.1001/jama.2016.17216.
    1. Hois J, Theofanou-Fuelbier D, Junk AJ. How to achieve explainability and transparency in human ai interaction. Commun. Comput. Inf. Sci. 2019 doi: 10.1007/978-3-030-23528-4_25.
    1. Naksuk N, et al. Right ventricular dysfunction and long-term risk of sudden cardiac death in patients with and without severe left ventricular dysfunction. Circ. Arrhythmia Electrophysiol. 2018 doi: 10.1161/CIRCEP.117.006091.
    1. Ali-Ahmed F, et al. Physicians’ perceptions of shared decision-making for implantable cardioverter-defibrillators: Results of a physician survey. J. Cardiovasc. Electrophysiol. 2019 doi: 10.1111/jce.14178.
    1. Schmidt A, et al. Infarct tissue heterogeneity by magnetic resonance imaging identifies enhanced cardiac arrhythmia susceptibility in patients with left ventricular dysfunction. Circulation. 2007;115:2006–2014. doi: 10.1161/CIRCULATIONAHA.106.653568.
    1. Arevalo HJ, et al. Arrhythmia risk stratification of patients after myocardial infarction using personalized heart models. Nat. Commun. 2016 doi: 10.1038/ncomms11437.
    1. Zhang Y, et al. Associations between scar characteristics by cardiac magnetic resonance and changes in left ventricular ejection fraction in primary prevention defibrillator recipients. Heart Rhythm. 2016 doi: 10.1016/j.hrthm.2016.04.013.
    1. Tao S, et al. Impaired left atrial function predicts inappropriate shocks in primary prevention implantable cardioverter-defibrillator candidates. J. Cardiovasc. Electrophysiol. 2017;28:796–805. doi: 10.1111/jce.13234.
    1. Okada DR, et al. Substrate spatial complexity analysis for the prediction of ventricular arrhythmias in patients with ischemic cardiomyopathy. Circ. Arrhythmia Electrophysiol. 2020 doi: 10.1161/CIRCEP.119.007975.
    1. Wu KC, et al. Baseline and dynamic risk predictors of appropriate implantable cardioverter defibrillator therapy. J. Am. Heart Assoc. 2020;9:e017002.
    1. Cox DR. Regression models and life-tables. J. R. Stat. Soc. Ser. B. 1972;34:187–202.
    1. Smith GCS, Seaman SR, Wood AM, Royston P, White IR. Correcting for optimistic prediction in small data sets. Am. J. Epidemiol. 2014;180:318–324. doi: 10.1093/aje/kwu140.
    1. Harrell FE, Califf RM, Pryor DB, Lee KL, Rosati RA. Evaluating the yield of medical tests. JAMA. 1982;247:2543–2546. doi: 10.1001/jama.1982.03320430047030.
    1. Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J. Am. Stat. Assoc. 1999 doi: 10.1080/01621459.1999.10474144.
    1. Leening MJG, Vedder MM, Witteman JCM, Pencina MJ, Steyerberg EW. Net reclassification improvement: Computation, interpretation, and controversies. Ann. Intern. Med. 2014 doi: 10.7326/m13-1522.

Source: PubMed

Sponsoren und Mitarbeiter

Krankheiten

Drogeninterventionen

3
Abonnieren