Top

Published in:

07-06-2023 | Heart Failure | Cardiac

Deep learning-based prognostic model using non-enhanced cardiac cine MRI for outcome prediction in patients with heart failure

Authors: Yifeng Gao, Zhen Zhou, Bing Zhang, Saidi Guo, Kairui Bo, Shuang Li, Nan Zhang, Hui Wang, Guang Yang, Heye Zhang, Tong Liu, Lei Xu

Published in: European Radiology | Issue 11/2023

Abstract

Objectives

To evaluate the performance of a deep learning-based multi-source model for survival prediction and risk stratification in patients with heart failure.

Methods

Patients with heart failure with reduced ejection fraction (HFrEF) who underwent cardiac magnetic resonance between January 2015 and April 2020 were retrospectively included in this study. Baseline electronic health record data, including clinical demographic information, laboratory data, and electrocardiographic information, were collected. Short-axis non-contrast cine images of the whole heart were acquired to estimate the cardiac function parameters and the motion features of the left ventricle. Model accuracy was evaluated using the Harrell’s concordance index. All patients were followed up for major adverse cardiac events (MACEs), and survival prediction was assessed using Kaplan–Meier curves.

Results

A total of 329 patients were evaluated (age 54 ± 14 years; men, 254) in this study. During a median follow-up period of 1041 days, 62 patients experienced MACEs and their median survival time was 495 days. When compared with conventional Cox hazard prediction models, deep learning models showed better survival prediction performance. Multi-data denoising autoencoder (DAE) model reached the concordance index of 0.8546 (95% CI: 0.7902–0.8883). Furthermore, when divided into phenogroups, the multi-data DAE model could significantly discriminate between the survival outcomes of the high-risk and low-risk groups compared with other models (p < 0.001).

Conclusions

The proposed deep learning (DL) model based on non-contrast cardiac cine magnetic resonance imaging could independently predict the outcome of patients with HFrEF and showed better prediction efficiency than conventional methods.

Clinical relevance statement

The proposed multi-source deep learning model based on cardiac magnetic resonance enables survival prediction in patients with heart failure.

Key Points

• A multi-source deep learning model based on non-contrast cardiovascular magnetic resonance (CMR) cine images was built to make robust survival prediction in patients with heart failure.

• The ground truth definition contains electronic health record data as well as DL-based motion data, and cardiac motion information is extracted by optical flow method from non-contrast CMR cine images.

• The DL-based model exhibits better prognostic value and stratification performance when compared with conventional prediction models and could aid in the risk stratification in patients with HF.

Available only for authorised users

Savarese G, Becher PM, Lund LH, Seferovic P, Rosano GMC, Coats A (2022) Global burden of heart failure: a comprehensive and updated review of epidemiology. Cardiovasc Res cvac013. https://doi.org/10.1093/cvr/cvac013

McDonagh TA, Metra M, Adamo M et al (2021) ESC Scientific Document Group. 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J 42(36):3599–3726. https://doi.org/10.1093/eurheartj/ehab368

Ambrosy AP, Fonarow GC, Butler J et al (2014) The global health and economic burden of hospitalizations for heart failure: lessons learned from hospitalized heart failure registries. J Am Coll Cardiol 63(12):1123–1133. https://doi.org/10.1016/j.jacc.2013.11.053CrossRefPubMed

Heidenreich PA, Bozkurt B, Aguilar D et al (2022) 2022 AHA/ACC/HFSA Guideline for the Management of Heart Failure: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation 145(18):e895–e1032. https://doi.org/10.1161/CIR.0000000000001063CrossRefPubMed

Khan SS, Ning H, Shah SJ et al (2019) 10-Year risk equations for incident heart failure in the general population. J Am Coll Cardiol 73:2388–2397. https://doi.org/10.1016/j.jacc.2019.02.057CrossRefPubMedPubMedCentral

Coughlin SS, Neaton JD, Sengupta A, Kuller LH (1994) Predictors of mortality from idiopathic dilated cardiomyopathy in 356,222 men screened for the Multiple Risk Factor Intervention Trial. Am J Epidemiol 139(2):166–172. https://doi.org/10.1093/oxfordjournals.aje.a116978CrossRefPubMed

Felker GM, Thompson RE, Hare JM et al (2000) Underlying causes and long-term survival in patients with initially unexplained cardiomyopathy. N Engl J Med 342:1077e84. https://doi.org/10.1056/NEJM200004133421502

Yancy CW, Jessup M, Bozkurt B et al (2017) 2017 ACC/AHA/HFSA Focused Update of the 2013 ACCF/AHA Guideline for the Management of Heart Failure: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Failure Society of America. J Am Coll Cardiol 70(6):776–803. https://doi.org/10.1016/j.jacc.2017.04.025CrossRefPubMed

Ponikowski P, Voors AA, Anker SD et al (2016) 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: The Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure Association (HFA) of the ESC. Eur J Heart Fail 18(8):891–975. https://doi.org/10.1002/ejhf.592

10.

Marrow BA, Cook SA, Prasad SK, McCann GP (2020) Emerging techniques for risk stratification in nonischemic dilated cardiomyopathy: JACC review topic of the week. J Am Coll Cardiol 75(10):1196–1207. https://doi.org/10.1016/j.jacc.2019.12.058CrossRefPubMed

11.

Segar MW, Vaduganathan M, Patel KV et al (2019) Machine learning to predict the risk of incident heart failure hospitalization among patients with diabetes: The WATCH-DM Risk Score. Diabetes Care 42:2298–2306. https://doi.org/10.2337/dc19-0587CrossRefPubMedPubMedCentral

12.

Grothues F, Smith GC, Moon JC et al (2002) Comparison of interstudy reproducibility of cardiovascular magnetic resonance with two-dimensional echocardiography in normal subjects and in patients with heart failure or left ventricular hypertrophy. Am J Cardiol 90(1):29–34. https://doi.org/10.1016/s0002-9149(02)02381-0CrossRefPubMed

13.

Patel AR, Kramer CM (2017) Role of cardiac magnetic resonance in the diagnosis and prognosis of nonischemic cardiomyopathy. JACC Cardiovasc Imaging 10(10 Pt A):1180–1193. https://doi.org/10.1016/j.jcmg.2017.08.005

14.

Peterzan MA, Rider OJ, Anderson LJ (2016) The role of cardiovascular magnetic resonance imaging in heart failure. Card Fail Rev 2(2):115–122. https://doi.org/10.15420/cfr.2016.2.2.115CrossRefPubMedPubMedCentral

15.

Becker MAJ, Cornel JH, van de Ven PM, van Rossum AC, Allaart CP, Germans T (2018) The prognostic value of late gadolinium-enhanced cardiac magnetic resonance imaging in nonischemic dilated cardiomyopathy: a review and meta-analysis. JACC Cardiovasc Imaging 11(9):1274–1284. https://doi.org/10.1016/j.jcmg.2018.03.006CrossRefPubMed

16.

Pezel T, Hovasse T, Sanguineti F et al (2021) Long-term prognostic value of stress CMR in patients with heart failure and preserved ejection fraction. JACC Cardiovasc Imaging 14(12):2319–2333. https://doi.org/10.1016/j.jcmg.2021.03.010CrossRefPubMed

17.

Puntmann VO, Carr-White G, Jabbour A et al (2016) T1-mapping and outcome in nonischemic cardiomyopathy: all-cause mortality and heart failure. JACC Cardiovasc Imaging 9(1):40–50. https://doi.org/10.1016/j.jcmg.2015.12.001CrossRefPubMed

18.

Romano S, Judd RM, Kim RJ et al (2018) Feature-tracking global longitudinal strain predicts death in a multicenter population of patients with ischemic and nonischemic dilated cardiomyopathy incremental to ejection fraction and late gadolinium enhancement. JACC Cardiovasc Imaging 11(10):1419–1429. https://doi.org/10.1016/j.jcmg.2017.10.024CrossRefPubMedPubMedCentral

19.

Pi SH, Kim SM, Choi JO et al (2018) Prognostic value of myocardial strain and late gadolinium enhancement on cardiovascular magnetic resonance imaging in patients with idiopathic dilated cardiomyopathy with moderate to severely reduced ejection fraction. J Cardiovasc Magn Reson 20(1):36. Published 2018 Jun 14. https://doi.org/10.1186/s12968-018-0466-7

20.

Haq IU, Haq I, Xu B (2021) Artificial intelligence in personalized cardiovascular medicine and cardiovascular imaging. Cardiovasc Diagn Ther 11(3):911–923. https://doi.org/10.21037/cdt.2020.03.09CrossRefPubMedPubMedCentral

21.

Yasmin F, Shah SMI, Naeem A et al (2021) Artificial intelligence in the diagnosis and detection of heart failure: the past, present, and future. Rev Cardiovasc Med 22(4):1095–1113. https://doi.org/10.31083/j.rcm2204121CrossRefPubMed

22.

Angraal S, Mortazavi BJ, Gupta A et al (2020) Machine learning prediction of mortality and hospitalization in heart failure with preserved ejection fraction. JACC Heart Fail 8(1):12–21. https://doi.org/10.1016/j.jchf.2019.06.013CrossRefPubMed

23.

Segar MW, Jaeger BC, Patel KV et al (2021) Development and validation of machine learning-based race-specific models to predict 10-year risk of heart failure: a multicohort analysis. Circulation 143(24):2370–2383. https://doi.org/10.1161/CIRCULATIONAHA.120.053134CrossRefPubMedPubMedCentral

24.

Leiner T, Rueckert D, Suinesiaputra A et al (2019) Machine learning in cardiovascular magnetic resonance: basic concepts and applications. J Cardiovasc Magn Reson 21(1):61. Published 2019 Oct 7. https://doi.org/10.1186/s12968-019-0575-y

25.

Patravali J, Jain S, Chilamkurthy S (2017) 2D-3D fully convolutional neural networks for cardiac MR segmentation. arXiv:1707.09813 [cs.CV]. https://doi.org/10.48550/arXiv.1707.09813

26.

Ronneberger O, Fischer P, Brox T (2015) U-net: convolutional networks for biomedical image segmentation. arXiv:1505.04597 [cs.CV]. https://doi.org/10.48550/arXiv.1505.04597

27.

Mortazavi BJ, Downing NS, Bucholz EM et al (2016) Analysis of machine learning techniques for heart failure readmissions. Circ Cardiovasc Qual Outcomes 9(6):629–640. https://doi.org/10.1161/CIRCOUTCOMES.116.003039CrossRefPubMedPubMedCentral

28.

Pandey A, Kagiyama N, Yanamala N et al (2021) Deep-learning models for the echocardiographic assessment of diastolic dysfunction. JACC Cardiovasc Imaging 14(10):1887–1900. https://doi.org/10.1016/j.jcmg.2021.04.010CrossRefPubMed

29.

Liang F, Xie W, Yu Y (2017) Beating heart motion accurate prediction method based on interactive multiple model: an information fusion approach. Biomed Res Int 2017:1279486. https://doi.org/10.1155/2017/1279486BelloGA,DawesTJW,DuanJ,etalCrossRefPubMedPubMedCentral

30.

Bello GA, Dawes TJW, Duan J et al (2019) Deep learning cardiac motion analysis for human survival prediction. Nat Mach Intell 1:95–104. https://doi.org/10.1038/s42256-019-0019-2CrossRefPubMedPubMedCentral

31.

Dawes TJW, de Marvao A, Shi W et al (2017) Machine learning of three-dimensional right ventricular motion enables outcome prediction in pulmonary hypertension: a cardiac MR imaging study. Radiology 283(2):381–390. https://doi.org/10.1148/radiol.2016161315CrossRefPubMed

32.

Guo S, Xu L, Feng C, Xiong H, Gao Z, Zhang H (2021) Multi-level semantic adaptation for few-shot segmentation on cardiac image sequences. Med Image Anal 73:102170. https://doi.org/10.1016/j.media.2021.102170

33.

Harrell FE Jr, Califf RM, Pryor DB, Lee KL, Rosati RA (1982) Evaluating the yield of medical tests. JAMA 247(18):2543–2546CrossRefPubMed

34.

Smith GC, Seaman SR, Wood AM, Royston P, White IR (2014) Correcting for optimistic prediction in small data sets. Am J Epidemiol 180(3):318–324. https://doi.org/10.1093/aje/kwu140CrossRefPubMedPubMedCentral

35.

Moons KG, Altman DG, Reitsma JB et al (2015) Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD): explanation and elaboration. Ann Intern Med 162(1):W1-73. https://doi.org/10.7326/M14-0698CrossRefPubMed

36.

Harrell FE Jr, Lee KL, Mark DB (1996) Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med 15(4):361–387. https://doi.org/10.1002/(SICI)1097-0258(19960229)15:4%3c361::AID-SIM168%3e3.0.CO;2-4CrossRefPubMed

37.

Rao RA, Jawaid O, Janish C, Raman SV (2021) When to use cardiovascular magnetic resonance in patients with heart failure. Heart Fail Clin 17(1):1–8. https://doi.org/10.1016/j.hfc.2020.09.001CrossRefPubMed

38.

Liu T, Gao Y, Wang H et al (2020) Association between right ventricular strain and outcomes in patients with dilated cardiomyopathy Heart 107(15):1233-1239. https://doi.org/10.1136/heartjnl-2020-317949

39.

Lehrke S, Lossnitzer D, Schöb M et al (2011) Use of cardiovascular magnetic resonance for risk stratification in chronic heart failure: prognostic value of late gadolinium enhancement in patients with non-ischaemic dilated cardiomyopathy. Heart 97(9):727–732. https://doi.org/10.1136/hrt.2010.205542CrossRefPubMed

40.

41.

Buss SJ, Breuninger K, Lehrke S et al (2015) Assessment of myocardial deformation with cardiac magnetic resonance strain imaging improves risk stratification in patients with dilated cardiomyopathy. Eur Heart J Cardiovasc Imaging 16(3):307–315. https://doi.org/10.1093/ehjci/jeu181CrossRefPubMed

42.

Litjens G, Ciompi F, Wolterink JM et al (2019) State-of-the-art deep learning in cardiovascular image analysis. JACC Cardiovasc Imaging 12(8 Pt 1):1549–1565. https://doi.org/10.1016/j.jcmg.2019.06.009CrossRefPubMed

43.

Chartrand G, Cheng PM, Vorontsov E et al (2017) Deep learning: a primer for radiologists. Radiographics 37(7):2113–2131. https://doi.org/10.1148/rg.2017170077CrossRefPubMed

44.

Shu S, Hong Z, Peng Q et al (2021) A machine-learning-based method to predict adverse events in patients with dilated cardiomyopathy and severely reduced ejection fractions. Br J Radiol 94(1127):20210259. https://doi.org/10.1259/bjr.20210259CrossRefPubMedPubMedCentral

45.

46.

Lee DS, Gona P, Vasan RS et al (2009) Relation of disease pathogenesis and risk factors to heart failure with preserved or reduced ejection fraction: insights from the framingham heart study of the national heart, lung, and blood institute. Circulation 119(24):3070–3077. https://doi.org/10.1161/CIRCULATIONAHA.108.815944CrossRefPubMedPubMedCentral

47.

Hogg K, Swedberg K, McMurray J (2004) Heart failure with preserved left ventricular systolic function; epidemiology, clinical characteristics, and prognosis. J Am Coll Cardiol 43(3):317–327. https://doi.org/10.1016/j.jacc.2003.07.046CrossRefPubMed

Title: Deep learning-based prognostic model using non-enhanced cardiac cine MRI for outcome prediction in patients with heart failure
Authors: Yifeng Gao
Zhen Zhou
Bing Zhang
Saidi Guo
Kairui Bo
Shuang Li
Nan Zhang
Hui Wang
Guang Yang
Heye Zhang
Tong Liu
Lei Xu
Publication date: 07-06-2023
Publisher: Springer Berlin Heidelberg
Keywords: Heart Failure
Magnetic Resonance Imaging
Magnetic Resonance Imaging
Published in: European Radiology / Issue 11/2023
Print ISSN: 0938-7994
Electronic ISSN: 1432-1084
DOI: https://doi.org/10.1007/s00330-023-09785-9

Keynote webinar | Spotlight on medication adherence

Springer Medicine

Deep learning-based prognostic model using non-enhanced cardiac cine MRI for outcome prediction in patients with heart failure

Abstract

Objectives

Methods

Results

Conclusions

Clinical relevance statement

Key Points

Keynote webinar | Spotlight on medication adherence

Springer Medicine

Abstract

Objectives

Methods

Results

Conclusions

Clinical relevance statement

Key Points

Please log in to get access to this content

Other articles of this Issue 11/2023

An automated surgical decision-making framework for partial or radical nephrectomy based on 3D-CT multi-level anatomical features in renal cell carcinoma

Arterial spin labelling MRI for brain tumour surveillance: do we really need cerebral blood flow maps?

MRI texture analysis of acetabular cancellous bone can discriminate between normal, cam positive, and cam-FAI hips

What predicts durable symptom relief of uterine fibroids treated with MRI-guided focused ultrasound? A multicenter trial in 8 academic centers

Automated CT quantification of interstitial lung abnormality and interstitial lung disease according to the Fleischner Society in patients with resectable lung cancer: prognostic significance

Post-mortem CT radiomics for the prediction of time since death