Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2023 Jan 2;13:3. doi: 10.1038/s41598-022-27211-w

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

© The Author(s) 2023

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

PMC Copyright notice

Performance of AIEchoDx. (a) Structure of feature extraction network; (a’) represents a heatmap of the features of one image extracted from the last hidden layer. (b) Structure of the diagnostic network; (b’) represents a heatmap of the feature matrix of a 45-frame echo video as an input of the diagnostic network. (c–g) The AUC (c), accuracy (d), sensitivity (e), specificity (f), and error-rate (g) values of the eightfold cross-validation results of the feature extraction network. (h) Confusion table displays the eightfold cross-validation results of the feature extraction network classified five categories from the test dataset 1. (i) The clustered heatmap illustrated 2048 features were extracted from the last hidden layer of the feature extraction network. 50 single frames from each category were selected from 50 patients for plotting the heatmap. (j–m) The frame number versus the values of matrix accuracy (j), F1 score (k), Cohen’s kappa (l), and AUC (m) of cardiac diseases and normal subjects from the diagnostic network. The diagnostic network reading a 45-frame video could achieve the best performance of the prediction in test dataset 1. (n,o) Receiver operating characteristic (ROC) curves of the 45-frame diagnostic model from test dataset 1 (The AUC values of ASD, DCM, HCM, prior MI, and Normal were 99.50%, 98.75%, 99.57%, 98.52%, and 98.70%, respectively). (p) The confusion matrices represent the prediction results of the five classifiers for test dataset 1.