Prediction of aortic valve regurgitation after continuous-flow left ventricular assist device implantation using artificial intelligence trained on acoustic spectra

Abstract

Significant aortic regurgitation (AR) is a common complication after continuous-flow left ventricular assist device (LVAD) implantation. Using machine-learning algorithms, this study was designed to examine valuable predictors obtained from LVAD sound and to provide models for identifying AR. During a 2-year follow-up period of 13 patients with Jarvik2000 LVAD, sound signals were serially obtained from the chest wall above the LVAD using an electronic stethoscope for 1 min at 40,000 Hz, and echocardiography was simultaneously performed to confirm the presence of AR. Among the 245 echocardiographic and acoustic data collected, we found 26 episodes of significant AR, which we categorized as “present”; the other 219 episodes were characterized as “none”. Wavelet (time–frequency) analysis was applied to the LVAD sound and 19 feature vectors of instantaneous spectral components were extracted. Important variables for predicting AR were searched using an iterative forward selection method. Seventy-five percent of 245 episodes were randomly assigned as training data and the remaining as test data. Supervised machine learning for predicting concomitant AR involved an ensemble classifier and tenfold stratified cross-validation. Of the 19 features, the most useful variables for predicting concomitant AR were the amplitude of the first harmonic, LVAD rotational speed during intermittent low speed (ILS), and the variation in the amplitude during normal rotation and ILS. The predictive accuracy and area under the curve were 91% and 0.73, respectively. Machine learning, trained on the time–frequency acoustic spectra, provides a novel modality for detecting concomitant AR during follow-up after LVAD.

Real-time regurgitation estimation in percutaneous left ventricular assist device fully supported condition using an unscented Kalman filter

“…In contrast, prior research (Yin et al 2023) leaned towards theoretical modeling and analysis. Regarding regurgitation signal acquisition, Yusuke et al (Misumi et al 2021) obtained signals by using an electronic stethoscope to continuously capture sound signals from the chest wall above the LVAD and confirm regurgitation through echocardiography. However, frequent ultrasonographic examinations are not practical in practice.…”

“…Yusuke et al (Misumi et al 2021) used acoustic signal analysis and machine learning methods to detect severe aortic regurgitation after LVAD installation. This time-frequency spectrum data-based machine learning algorithm provides a novel method for identifying concurrent aortic regurgitation during LVAD follow-up.…”

Artificial intelligence guidance of advanced heart failure therapies: A systematic scoping review

“…The process is repeated until the algorithm is tested on all folds, and the average performance across all test folds is reported (44). Three studies, in which sample size was less than 60, used leave-one-out crossvalidation in evaluating the model's performance; evaluating the model on one instance / case and training the model using the rest of the cases, iteratively (32,34,37). External validation was only used in…”

Degenerative changes of the aortic valve during left ventricular assist device support

“…Today, the annual number of patients receiving LVAD has surpassed the corresponding numbers for cardiac transplantation in most countries 2,3 . The development of aortic valve regurgitation due to structural alterations of the aortic valve is a frequent phenomenon complicating long‐term LVAD therapy 4, 5 . Reoperation on the aortic valve after previous LVAD implantation is expected to have a relevant negative impact on the outcome of affected patients.…”

“… 2 , 3 The development of aortic valve regurgitation due to structural alterations of the aortic valve is a frequent phenomenon complicating long‐term LVAD therapy. 4 , 5 Reoperation on the aortic valve after previous LVAD implantation is expected to have a relevant negative impact on the outcome of affected patients. Furthermore, aortic valves of LVAD patients eventually receiving heart transplantation (HTx) are regularly harvested for homografts.…”