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

Abstract

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.

Efficient detection of aortic stenosis using morphological characteristics of cardiomechanical signals and heart rate variability parameters

“…Figure 1 depicts the ECG, , and from top to bottom, respectively, wherein and are annotated according to the method discussed in the following sections. These axes of GCG provide useful information about the cardiac activity timing intervals as demonstrated in previous research works 21 , 22 . Hence, they are expected to provide potential insights into the diagnosis of AS.…”

“…More recently, fetal heart rate (f-HR) has been extracted using SCG and GCG modalities, where promising results were achieved in comparison with concurrently-recorded fetal cardiotocography 23 . Deploying machine learning (ML) algorithms, our research group has targeted AS detection based on the SCG/GCG technology 21 , 22 . In these works, the time-frequency representation of all ten second SCG/GCG segments were generated, out of which features such as the energy of frequency bands were extracted.…”

“…Predictive models were then trained to classify between AS and non-AS subjects. Yet, a major drawback of the aforementioned works is the need for a reference signal, e.g., ECG 21 or PPG 22 , for either signal segmentation or feature extraction. Moreover, there is no physical meaning behind the frequency-band energies which were used as features.…”

An Open-Access Database for the Evaluation of Cardio-Mechanical Signals From Patients With Valvular Heart Diseases

“…The second cohort consisted of nine AS patients who needed transcatheter aortic valve replacements. Data collection was completed in 2019, and the presentation of the corresponding study was in Yang et al (2020b). All the data were collected either in the cardiac care unit or the pre-operative room of the CUMC, in collaboration with Dr. Philip Green.…”

Transfer learning for non-image data in clinical research: a scoping review

“…Moreover, transfer learning was used for prediction of glucose levels [26,27], estimation of Parkinson's disease severity [71], detection of cognitive impairment [33] and schizophrenia [52], and forecasting of infectious disease trends [28] and outbreaks [29], among other applications [72][73][74][75][76][77][78][79][80][81][82].…”