Journal DOI 10.68956/journal.human-performance-futures

Creative Commons source reading · Prediction and forecasting

Artificial intelligence-based prediction of procedural success in transcatheter mitral valve edge-to-edge repair

Mathias Claeys; Neil Fam; Xiaojiao Xiao; Tejas Vyas; Mohsena Chowdhury; Gordon Moe; Sami Alnasser; Patrick Prud’homme; Anas Merdad; Akshay Bagai; Kim A. Connelly; Jeremy Edwards; Guanghui Wang; Géraldine Ong

Originally published in Echo Res Pract · 2026-09-07

From the article

Selected source text. JournalOS removes markup and shortens excerpts where indicated. The full article provides the complete context.

Abstract

BackgroundPatient selection for transcatheter mitral valve edge-to-edge repair (MTEER) remains challenging, particularly in individuals with complex mitral valve anatomy. Conventional risk scores incorporate limited clinical variables and do not adequately account for detailed echocardiographic features, resulting in suboptimal prediction of procedural success.ObjectivesTo develop and evaluate machine learning (ML) and deep learning (DL) models integrating clinical and echocardiographic data to predict procedural success following MTEER.MethodsConsecutive patients undergoing MTEER at a single tertiary center between 2014 and 2022 were retrospectively analyzed. Procedural success was defined as residual mitral regurgitation ≤ mild and mean mitral valve gradient < 5 mmHg at the end of the procedure. Pre-procedural transesophageal…

Discussion

In this study, artificial intelligence–based approaches were feasible for predicting procedural success following MTEER. Machine learning models based on structured echocardiographic measurements demonstrated modest predictive performance, whereas DL models derived from pre-procedural TEE imaging achieved superior discrimination and outperformed conventional logistic regression. These findings suggest that image-based deep learning approaches may capture complex echocardiographic features that are not adequately represented by structured echocardiographic measurements alone. As MTEER has become a well-established, guideline-directed therapy for patients with degenerative and functional MR who are at elevated surgical risk, accurate prediction of procedural success is increasingly important in contemporary clinical practice [3]. In…

Conclusion

In conclusion, AI-based approaches, specifically DL models, were feasible for predicting procedural success following MTEER. Deep-learning models derived directly from pre-procedural TEE imaging achieved superior discrimination and outperformed conventional logistic regression. These findings suggest that important anatomical and functional information relevant to MTEER outcomes may be embedded within imaging data and not fully captured by conventional echocardiographic measurements. Further validation in larger multicenter cohorts is warranted to determine the clinical utility of artificial intelligence–based tools for patient selection, procedural planning, and risk stratification in contemporary MTEER practice. Table 1Baseline clinical characteristicsCohort (n = 467)Age, years77 ± 11Sex female, n (%)209 (45)Diabetes, n (%)118…

Editorial reading guide

  • What data created and tested the model, and was evaluation independent?
  • What is the cost of a false positive, false negative or out-of-distribution decision?