Deep Learning from Atrioventricular Plane Displacement in Patients with Takotsubo Syndrome: Lighting Up the Black-box

Overview

Journal Eur Heart J Digit Health

Publisher Oxford University Press

Specialty Cardiology & Vascular Diseases

Date 2024 Mar 20

PMID 38505490

Authors

Fahim Zaman

Nicholas Isom

Amanda Chang

Yi Grace Wang

Ahmed Abdelhamid

Arooj Khan

Majesh Makan

Mahmoud Abdelghany

Xiaodong Wu

Kan Liu

Affiliations

Soon will be listed here.

Abstract

Aims: The spatiotemporal deep convolutional neural network (DCNN) helps reduce echocardiographic readers' erroneous 'judgement calls' on Takotsubo syndrome (TTS). The aim of this study was to improve the interpretability of the spatiotemporal DCNN to discover latent imaging features associated with causative TTS pathophysiology.

Methods And Results: We applied gradient-weighted class activation mapping analysis to visualize an established spatiotemporal DCNN based on the echocardiographic videos to differentiate TTS (150 patients) from anterior wall ST-segment elevation myocardial infarction (STEMI, 150 patients). Forty-eight human expert readers interpreted the same echocardiographic videos and prioritized the regions of interest on myocardium for the differentiation. Based on visualization results, we completed optical flow measurement, myocardial strain, and Doppler/tissue Doppler echocardiography studies to investigate regional myocardial temporal dynamics and diastology. While human readers' visualization predominantly focused on the apex of the heart in TTS patients, the DCNN temporal arm's saliency visualization was attentive on the base of the heart, particularly at the atrioventricular (AV) plane. Compared with STEMI patients, TTS patients consistently showed weaker peak longitudinal displacement (in pixels) in the basal inferoseptal (systolic: 2.15 ± 1.41 vs. 3.10 ± 1.66, < 0.001; diastolic: 2.36 ± 1.71 vs. 2.97 ± 1.69, = 0.004) and basal anterolateral (systolic: 2.70 ± 1.96 vs. 3.44 ± 2.13, = 0.003; diastolic: 2.73 ± 1.70 vs. 3.45 ± 2.20, = 0.002) segments, and worse longitudinal myocardial strain in the basal inferoseptal (-8.5 ± 3.8% vs. -9.9 ± 4.1%, = 0.013) and basal anterolateral (-8.6 ± 4.2% vs. -10.4 ± 4.1%, = 0.006) segments. Meanwhile, TTS patients showed worse diastolic mechanics than STEMI patients (E'/septal: 5.1 ± 1.2 cm/s vs. 6.3 ± 1.5 cm/s, < 0.001; S'/septal: 5.8 ± 1.3 cm/s vs. 6.8 ± 1.4 cm/s, < 0.001; E'/lateral: 6.0 ± 1.4 cm/s vs. 7.9 ± 1.6 cm/s, < 0.001; S'/lateral: 6.3 ± 1.4 cm/s vs. 7.3 ± 1.5 cm/s, < 0.001; E/E': 15.5 ± 5.6 vs. 12.5 ± 3.5, < 0.001).

Conclusion: The spatiotemporal DCNN saliency visualization helps identify the pattern of myocardial temporal dynamics and navigates the quantification of regional myocardial mechanics. Reduced AV plane displacement in TTS patients likely correlates with impaired diastolic mechanics.

References

Alam M . The atrioventricular plane displacement as a means of evaluating left ventricular systolic function in acute myocardial infarction. Clin Cardiol. 1991; 14(7):588-94. DOI: 10.1002/clc.4960140711. View

Boynton G . Attention and visual perception. Curr Opin Neurobiol. 2005; 15(4):465-9. DOI: 10.1016/j.conb.2005.06.009. View

Ghadri J, Kato K, Cammann V, Gili S, Jurisic S, Di Vece D . Long-Term Prognosis of Patients With Takotsubo Syndrome. J Am Coll Cardiol. 2018; 72(8):874-882. DOI: 10.1016/j.jacc.2018.06.016. View

Laumer F, Di Vece D, Cammann V, Wurdinger M, Petkova V, Schonberger M . Assessment of Artificial Intelligence in Echocardiography Diagnostics in Differentiating Takotsubo Syndrome From Myocardial Infarction. JAMA Cardiol. 2022; 7(5):494-503. PMC: 8968683. DOI: 10.1001/jamacardio.2022.0183. View

Obermeyer Z, Emanuel E . Predicting the Future - Big Data, Machine Learning, and Clinical Medicine. N Engl J Med. 2016; 375(13):1216-9. PMC: 5070532. DOI: 10.1056/NEJMp1606181. View

Betancur J, Otaki Y, Motwani M, Fish M, Lemley M, Dey D . Prognostic Value of Combined Clinical and Myocardial Perfusion Imaging Data Using Machine Learning. JACC Cardiovasc Imaging. 2017; 11(7):1000-1009. PMC: 5902657. DOI: 10.1016/j.jcmg.2017.07.024. View

Litjens G, Ciompi F, Wolterink J, de Vos B, Leiner T, Teuwen J . State-of-the-Art Deep Learning in Cardiovascular Image Analysis. JACC Cardiovasc Imaging. 2019; 12(8 Pt 1):1549-1565. DOI: 10.1016/j.jcmg.2019.06.009. View

Kranidis A, Kostopoulos K, Kappos K, Anthopoulos L . Diastolic mitral annular motion: assessment of left ventricular filling. Eur Heart J. 1996; 17(10):1606-7. DOI: 10.1093/oxfordjournals.eurheartj.a014736. View

Hughes J, Yuan N, He B, Ouyang J, Ebinger J, Botting P . Deep learning evaluation of biomarkers from echocardiogram videos. EBioMedicine. 2021; 73:103613. PMC: 8524103. DOI: 10.1016/j.ebiom.2021.103613. View

10.

Singh T, Khan H, Gamble D, Scally C, Newby D, Dawson D . Takotsubo Syndrome: Pathophysiology, Emerging Concepts, and Clinical Implications. Circulation. 2022; 145(13):1002-1019. PMC: 7612566. DOI: 10.1161/CIRCULATIONAHA.121.055854. View

11.

Ouyang D, He B, Ghorbani A, Yuan N, Ebinger J, Langlotz C . Video-based AI for beat-to-beat assessment of cardiac function. Nature. 2020; 580(7802):252-256. PMC: 8979576. DOI: 10.1038/s41586-020-2145-8. View

12.

Lang R, Badano L, Mor-Avi V, Afilalo J, Armstrong A, Ernande L . Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr. 2015; 28(1):1-39.e14. DOI: 10.1016/j.echo.2014.10.003. View

13.

Vafaeezadeh M, Behnam H, Hosseinsabet A, Gifani P . Automatic morphological classification of mitral valve diseases in echocardiographic images based on explainable deep learning methods. Int J Comput Assist Radiol Surg. 2021; 17(2):413-425. DOI: 10.1007/s11548-021-02542-7. View

14.

Lei J, Chen J, Dogra M, Gebska M, Shetty S, Ponnapureddy R . "Takotsubo effect" in patients with ST segment elevation myocardial infarction. Eur Heart J Acute Cardiovasc Care. 2020; 9(7):711-720. DOI: 10.1177/2048872620926680. View

15.

Ghadri J, Wittstein I, Prasad A, Sharkey S, Dote K, Akashi Y . International Expert Consensus Document on Takotsubo Syndrome (Part I): Clinical Characteristics, Diagnostic Criteria, and Pathophysiology. Eur Heart J. 2018; 39(22):2032-2046. PMC: 5991216. DOI: 10.1093/eurheartj/ehy076. View

16.

Schnuerch R, Kreitz C, Gibbons H, Memmert D . Not quite so blind: Semantic processing despite inattentional blindness. J Exp Psychol Hum Percept Perform. 2016; 42(4):459-63. DOI: 10.1037/xhp0000205. View

17.

Beam A, Kohane I . Big Data and Machine Learning in Health Care. JAMA. 2018; 319(13):1317-1318. DOI: 10.1001/jama.2017.18391. View

18.

Lorenz-Spreen P, Monsted B, Hovel P, Lehmann S . Accelerating dynamics of collective attention. Nat Commun. 2019; 10(1):1759. PMC: 6465266. DOI: 10.1038/s41467-019-09311-w. View

19.

Cerna A, Jing L, Good C, vanMaanen D, Raghunath S, Suever J . Deep-learning-assisted analysis of echocardiographic videos improves predictions of all-cause mortality. Nat Biomed Eng. 2021; 5(6):546-554. DOI: 10.1038/s41551-020-00667-9. View

20.

Elgendy A, Elgendy I, Mansoor H, Mahmoud A . Clinical presentations and outcomes of Takotsubo syndrome in the setting of subarachnoid hemorrhage: A systematic review and meta-analysis. Eur Heart J Acute Cardiovasc Care. 2016; 7(3):236-245. DOI: 10.1177/2048872616679792. View