Dataset Information

Predicting post-operative right ventricular failure using video-based deep learning.

ABSTRACT: Despite progressive improvements over the decades, the rich temporally resolved data in an echocardiogram remain underutilized. Human assessments reduce the complex patterns of cardiac wall motion, to a small list of measurements of heart function. All modern echocardiography artificial intelligence (AI) systems are similarly limited by design - automating measurements of the same reductionist metrics rather than utilizing the embedded wealth of data. This underutilization is most evident where clinical decision making is guided by subjective assessments of disease acuity. Predicting the likelihood of developing post-operative right ventricular failure (RV failure) in the setting of mechanical circulatory support is one such example. Here we describe a video AI system trained to predict post-operative RV failure using the full spatiotemporal density of information in pre-operative echocardiography. We achieve an AUC of 0.729, and show that this ML system significantly outperforms a team of human experts at the same task on independent evaluation.

SUBMITTER: Shad R

PROVIDER: S-EPMC8408163 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Right ventricular impairment is a predictor of cardiovascular outcomes in patients with degenerative mitral regurgitation. However, the time course of right ventricular functional changes post-surgical mitral valve repair remains largely unknown. Herein, using right ventricular-focused echocardiography, we aimed to investigate right ventricular reserve and its impact on hospitalization for heart failure after mitral valve repair. In this prospective study, we enrolled 108 patients scheduled to undergo surgical repair of degenerative mitral regurgitation. Echocardiography, including right ventricular strain analysis, was performed prior to, and one month and six months post mitral valve repair. Right ventricular strain that improved one month post-surgery was defined as reserved right ventricular. In addition, any cardiovascular outcomes comprising heart failure that required admission were recorded. The median follow-up duration is 31 months. Despite a significant improvement in mitral valve regurgitant volume post-operatively, left ventricular ejection fraction (LVEF) at six months was similar to LVEF at baseline. There was a transient decrease in LV longitudinal strain at one month that was recovered six months post mitral valve repair. Regarding the right ventricular, in contrast with conventional right ventricular parameters, including right ventricular tissue Doppler S', fractional area change and tricuspid annular plane systolic excursion (TAPSE), only resolution of right ventricular strain at one month predicted the subsequent myocardial recovery. Furthermore, patients with reserved right ventricular had a lower risk of hospitalization for heart failure compared to those with non-reserved right ventricular. Collectively, the early resolution of right ventricular strain is associated with the improvement in right ventricular function (measured by TAPSE) and in heart failure hospitalization in patients who had undergone surgical mitral valve repair for degenerative mitral regurgitation.

Dataset Information

Predicting post-operative right ventricular failure using video-based deep learning.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets