Classification of Aortic Stenosis Before and After Transcatheter Aortic Valve Replacement Using Cardio-mechanical Modalities.

This paper reports our study on the impact of transcatheter aortic valve replacement (TAVR) on the classification of aortic stenosis (AS) patients using cardio-mechanical modalities. Machine learning algorithms such as decision tree, random forest, and neural network were applied to conduct two tasks. Firstly, the pre- and post-TAVR data are evaluated with the classifiers trained in the literature. Secondly, new classifiers are trained to classify between pre- and post-TAVR data. Using analysis of variance, the features that are significantly different between pre- and post-TAVR patients are selected and compared to the features used in the pre-trained classifiers. The results suggest that pre-TAVR subjects could be classified as AS patients but post-TAVR could not be classified as healthy subjects. The features which differentiate pre- and post-TAVR patients reveal different distributions compared to the features that classify AS patients and healthy subjects. These results could guide future work in the classification of AS as well as the evaluation of the recovery status of patients after TAVR treatment.

Classification of aortic stenosis using conventional machine learning and deep learning methods based on multi-dimensional cardio-mechanical signals.

This paper introduces a study on the classification of aortic stenosis (AS) based on cardio-mechanical signals collected using non-invasive wearable inertial sensors. Measurements were taken from 21 AS patients and 13 non-AS subjects. A feature analysis framework utilizing Elastic Net was implemented to reduce the features generated by continuous wavelet transform (CWT). Performance comparisons were conducted among several machine learning (ML) algorithms, including decision tree, random forest, multi-layer perceptron neural network, and extreme gradient boosting. In addition, a two-dimensional convolutional neural network (2D-CNN) was developed using the CWT coefficients as images. The 2D-CNN was made with a custom-built architecture and a CNN based on Mobile Net via transfer learning. After the reduction of features by 95.47%, the results obtained report 0.87 on accuracy by decision tree, 0.96 by random forest, 0.91 by simple neural network, and 0.95 by XGBoost. Via the 2D-CNN framework, the transfer learning of Mobile Net shows an accuracy of 0.91, while the custom-constructed classifier reveals an accuracy of 0.89. Our results validate the effectiveness of the feature selection and classification framework. They also show a promising potential for the implementation of deep learning tools on the classification of AS.