Please use this identifier to cite or link to this item:
http://hdl.handle.net/1942/43621
Title: | Arrhythmic Mitral Valve Prolapse Phenotype: An Unsupervised Machine Learning Analysis Using a Multicenter Cardiac MRI Registry | Authors: | Akyea, Ralph Kwame Figliozzi, Stefano Lopes, Pedro M. Bauer, Klemens B. MOURA FERREIRA, Sara Tondi, Lara Mushtaq, Saima Censi, Stefano Pavon, Anna Giulia Bassi, Ilaria Galian-Gay, Laura Teske, Arco J. Biondi, Federico Filomena, Domenico Stylianidis, Vasileios Torlasco, Camilla Muraru, Denisa Monney, Pierre Quattrocchi, Giuseppina Maestrini, Viviana Agati, Luciano Monti, Lorenzo Pedrotti, Patrizia Vandenberk, Bert Squeri, Angelo Lombardi, Massimo Ferreira, Antonio M. Schwitter, Juerg Aquaro, Giovanni Donato Pontone, Gianluca Chiribiri, Amedeo Palomares, Jose F. Rodriguez Yilmaz, Ali Andreini, Daniele Florian, Anca-Rezeda Francone, Marco Leiner, Tim Abecasis, Joao Badano, Luigi Paolo Bogaert, Jan Georgiopoulos, Georgios Masci, Pier-Giorgio |
Issue Date: | 2024 | Publisher: | RADIOLOGICAL SOC NORTH AMERICA (RSNA) | Source: | Radiology-Cardiothoracic Imaging, 6 (3) (Art N° e230247) | Abstract: | Purpose: To use unsupervised machine learning to identify phenotypic clusters with increased risk of arrhythmic mitral valve prolapse (MVP). Materials and Methods: This retrospective study included patients with MVP without hemodynamically significant mitral regurgitation or left ventricular (LV) dysfunction undergoing late gadolinium enhancement (LGE) cardiac MRI between October 2007 and June 2020 in 15 European tertiary centers. The study end point was a composite of sustained ventricular tachycardia, (aborted) sudden cardiac death, or unexplained syncope. Unsupervised data-driven hierarchical k-mean algorithm was utilized to identify phenotypic clusters. The association between clusters and the study end point was assessed by Cox proportional hazards model. Results: A total of 474 patients (mean age, 47 years +/- 16 [SD]; 244 female, 230 male) with two phenotypic clusters were identified. Patients in cluster 2 (199 of 474, 42%) had more severe mitral valve degeneration (ie, bileaflet MVP and leaflet displacement), left and right heart chamber remodeling, and myocardial fibrosis as assessed with LGE cardiac MRI than those in cluster 1. Demographic and clinical features (ie, symptoms, arrhythmias at Holter monitoring) had negligible contribution in differentiating the two clusters. Compared with cluster 1, the risk of developing the study end point over a median follow-up of 39 months was significantly higher in cluster 2 patients (hazard ratio: 3.79 [95% CI: 1.19, 12.12], P = .02) after adjustment for LGE extent. Conclusion: Among patients with MVP without significant mitral regurgitation or LV dysfunction, unsupervised machine learning enabled the identification of two phenotypic clusters with distinct arrhythmic outcomes based primarily on cardiac MRI features. These results encourage the use of in-depth imaging-based phenotyping for implementing arrhythmic risk prediction in MVP. | Notes: | Masci, PG (corresponding author), Kings Coll London, Fac Life Sci & Med, Sch Biomed Engn & Imaging Sci, Westminster Bridge Rd, London SE1 7EH, England. pier_giorgio.masci@kcl.ac.uk |
Keywords: | MR Imaging;Cardiac;Cardiac MRI;Mitral Valve Prolapse;Cluster Analysis;Ventricular Arrhythmia;Sudden Cardiac Death;Unsupervised Machine Learning | Document URI: | http://hdl.handle.net/1942/43621 | ISSN: | 2638-6135 | e-ISSN: | 2638-6135 | DOI: | 10.1148/ryct.230247 | ISI #: | 001272579300019 | Rights: | RSNA, 2024 | Category: | A1 | Type: | Journal Contribution |
Appears in Collections: | Research publications |
Show full item record
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.