Predicting Mitral Valve mTEER Surgery Outcomes Using Machine Learning and Deep Learning Techniques
Tejas Vyas, Mohsena Chowdhury, Xiaojiao Xiao, Mathias Claeys, Géraldine Ong, Guanghui Wang
TL;DR
The study tackles predicting mitral valve mTEER surgery outcomes using machine learning and deep learning on a novel dataset of 467 patients with labeled echocardiogram videos and Transesophageal Echocardiography measurements. It benchmarks six ML algorithms and two CNN-based DL models (VGG16 and EfficientNet-B0) under a rigorous 10-fold, patient-level cross-validation, establishing baseline performance for this task. Logistic Regression emerges as the strongest ML model and VGG16 as the strongest DL model, with DL generally achieving higher frame-level accuracy and ML delivering competitive AUC and F1 scores given the limited data. This work demonstrates feasibility for imaging- and report-based prediction in mTEER and lays groundwork for future multimodal, multi-class approaches to enhance clinical decision-making and outcome guidance.
Abstract
Mitral Transcatheter Edge-to-Edge Repair (mTEER) is a medical procedure utilized for the treatment of mitral valve disorders. However, predicting the outcome of the procedure poses a significant challenge. This paper makes the first attempt to harness classical machine learning (ML) and deep learning (DL) techniques for predicting mitral valve mTEER surgery outcomes. To achieve this, we compiled a dataset from 467 patients, encompassing labeled echocardiogram videos and patient reports containing Transesophageal Echocardiography (TEE) measurements detailing Mitral Valve Repair (MVR) treatment outcomes. Leveraging this dataset, we conducted a benchmark evaluation of six ML algorithms and two DL models. The results underscore the potential of ML and DL in predicting mTEER surgery outcomes, providing insight for future investigation and advancements in this domain.