Evaluation of an artificial intelligence-based system for echocardiographic estimation of right atrial pressure

Figures

• 

  Fig. 1. Automated pipeline for echocardiography assessment of dIVC, cIVC, and RAP estimation

• 

  Fig. 2. Left: example of good quality IVC view retrieved automatically from a set of other views; right: example of IVC view retrieved automatically from a set of other views and labeled as unusable (bad quality) as the IVC’s boundary is not clear

• 

  Fig. 3. Left: the automatically generated mask overlayed on the original image; right: the contour of the overlayed mask

• 

  Fig. 4. Top left: automated dIVC in frame 253 of a given video with a dIVC of 1.73 cm (maximum dIVC); top right: automated dIVC in frame 183 of a given video with a dIVC of 0.475 cm (minimum dIVC); bottom: dIVC in each frame of a given clip, where the peaks indicate maximum dIVC and valleys indicate minimum dIVC

• 

  Fig. 5. Correlation and B &A plots between automated and human dIVC (at 2 cm location)

• 

  Fig. 6. First row: non-normalized confusion matrices for automated RAP estimation using ASE Criterion (left) and NIH Criterion (right). Second row: normalized confusion matrices for automated RAP estimation using ASE Criterion (left) and NIH Criterion (right). The normalized confusion matrix is generated by dividing each value in the confusion matrix by the sum of the corresponding row

• 

  Fig. 7. Overview of IVC retrieval (quality assessment and view classification) stage. Each block represents MobileNetV2-s residual blocks. GAP and FC represent the global average pooling layer and fully connected layer, respectively. OpenMax is a replacement for the SoftMax function that enables open-world learning

• 

  Fig. 8. TaNet for IVC region segmentation. TaNet has two main components: STN for IVC region localization and segmentation with 3 pathways, spatial (detail) path (SP), handcrafted path (HP), global or context path (GP). STN focuses the segmentation attention on the cardiac region. The numbers in the segmentation cubes are the size ratios to the resolution of the input. L(z), \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\theta$$\end{document}θ, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\theta (G)$$\end{document}θ(G) are STN parameters