Using Artificial Intelligence for Rheumatic Heart Disease Detection by Echocardiography: Focus on Mitral Regurgitation.

BACKGROUND: Identification of children with latent rheumatic heart disease (RHD) by echocardiography, before onset of symptoms, provides an opportunity to initiate secondary prophylaxis and prevent disease progression. There have been limited artificial intelligence studies published assessing the potential of machine learning to detect and analyze mitral regurgitation or to detect the presence of RHD on standard portable echocardiograms. METHODS AND RESULTS: We used 511 echocardiograms in children, focusing on color Doppler images of the mitral valve. Echocardiograms were independently reviewed by an expert adjudication panel. Among 511 cases, 229 were normal, and 282 had RHD. Our automated method included harmonization of echocardiograms to localize the left atrium during systole using convolutional neural networks and RHD detection using mitral regurgitation jet analysis and deep learning models with an attention mechanism. We identified the correct view with an average accuracy of 0.99 and the correct systolic frame with an average accuracy of 0.94 (apical) and 0.93 (parasternal long axis). It localized the left atrium with an average Dice coefficient of 0.88 (apical) and 0.9 (parasternal long axis). Maximum mitral regurgitation jet measurements were similar to expert manual measurements (P value=0.83) and a 9-feature mitral regurgitation analysis showed an area under the receiver operating characteristics curve of 0.93, precision of 0.83, recall of 0.92, and F1 score of 0.87. Our deep learning model showed an area under the receiver operating characteristics curve of 0.84, precision of 0.78, recall of 0.98, and F1 score of 0.87. CONCLUSIONS: Artificial intelligence has the potential to detect RHD as accurately as expert cardiologists and to improve with more data. These innovative approaches hold promise to scale echocardiography screening for RHD.

Frequency-Based Maternal Electrocardiogram Attenuation for Fetal Electrocardiogram Analysis.

Fetal electrocardiogram (ECG) waveform analysis along with cardiac time intervals (CTIs) measurements are critical for the management of high-risk pregnancies. Currently, there is no system that can consistently and accurately measure fetal ECG. In this work, we present a new automatic approach to attenuate the maternal ECG in the frequency domain and enhance it with measurable CTIs. First, the coherent components between the maternal ECG and abdominal ECG were identified and subtracted from the latter in the frequency domain. The residual was then converted into the time domain using the inverse Fourier transform to yield the fetal ECG. This process was improved by averaging multiple beats. Two fetal cardiologists, blinded to the method, assessed the quality of fetal ECG based on a grading system and measured the CTIs. We evaluated the fetal ECG quality of our method and time-based methods using one synthetic dataset, one human dataset available in the public domain, and 37 clinical datasets. Among the 37 datasets analyzed, the mean average (± standard deviation) grade was 3.49 ± 1.22 for our method vs. 2.64 ± 1.26 for adaptive interference cancellation (p-value < 0.001), thus showing the frequency-based fetal ECG extraction was the superior method, as assessed from our clinicians' perspectives. This method has the potential for use in clinical settings.