Abstract:
Rheumatic heart disease (RHD) is a common condition in young children living in low- and middle-income countries where cardiology expertise is insufficiently available. M...Show MoreMetadata
Abstract:
Rheumatic heart disease (RHD) is a common condition in young children living in low- and middle-income countries where cardiology expertise is insufficiently available. Mitral regurgitation (MR) is an important imaging feature used to diagnose RHD. We present a new, automatic framework to detect and characterize the MR jet on color Doppler echocardiograms, followed by RHD detection. Our method includes (1) echocardiogram harmonization, (2) localization and characterization of the maximum MR jet, and (3) RHD detection. We evaluated the performance using 1,807 color Doppler echocardiograms from 511 children with MR on two acquisition views (parasternal and apical). We localized the maximum MR jet and measured the length with an accuracy similar to that of expert manual measurements (p-value = 0.83), resulting in an average accuracy of 0.86, sensitivity of 0.92, and specificity of 0.79 for RHD detection. Our automatic approach has the potential to detect RHD as reliably as expert clinicians.
Date of Conference: 18-21 April 2023
Date Added to IEEE Xplore: 01 September 2023
ISBN Information: