Related Experiment Video
Updated: Feb 1, 2026
Deep Learning-Based Segmentation of Cryo-Electron Tomograms
Published on: November 11, 2022
Deep learning for predicting in-hospital mortality among heart disease patients based on echocardiography
Joon-Myoung Kwon1, Kyung-Hee Kim2, Ki-Hyun Jeon2
1Department of Emergency Medicine, Mediplex Sejong Hospital, Incheon, Korea.
Insights
A new deep learning (DL) model using echocardiography accurately predicts heart disease (HD) mortality. This advanced model outperforms existing methods, offering a valuable tool for patient screening and initial treatment decisions.
Area of Science:
- Cardiology
- Artificial Intelligence
- Medical Informatics
Background:
- Heart disease (HD) is a leading global cause of mortality.
- Existing HD mortality prediction models have limitations in early screening and initial treatment guidance.
- There is a need for accurate and accessible prediction tools for critically-ill HD patients.
Purpose of the Study:
- To develop and validate a deep learning (DL) based mortality prediction model for heart disease (HD) using echocardiography data.
- To assess the model's performance in predicting in-hospital mortality among adult HD patients.
- To compare the DL model's accuracy against established clinical prediction scores.
Main Methods:
- A multicenter retrospective cohort study involving adult HD patients who underwent echocardiography.
- Extraction of predictor variables from echocardiography reports utilizing text mining.
- Development of a DL prediction model using data from Hospital A and external validation with data from Hospital B.
- Subgroup analysis for coronary heart disease (CHD) and heart failure (HF) patients, comparing DL model performance with GRACE, TIMI, MAGGIC, and GWTG-HF scores.
Main Results:
- The study included 25,776 patients with 1026 mortalities.
- The DL model achieved high areas under the receiver operating characteristic curve (AUROC): 0.912 (internal validation), 0.898 (external validation), 0.958 (CHD), and 0.913 (HF).
- The DL model significantly outperformed existing prediction models (GRACE, TIMI, MAGGIC, GWTG-HF) in predicting in-hospital mortality.
Conclusions:
- An echocardiography-based DL model demonstrates superior accuracy in predicting in-hospital mortality for HD patients compared to current models.
- This DL model offers a promising advancement for early risk stratification and guiding clinical decisions in HD management.
- The findings highlight the potential of AI and text mining in leveraging echocardiography data for improved cardiovascular outcome prediction.
Background:
Heart disease (HD) is the leading cause of global death; there are several mortality prediction models of HD for identifying critically-ill patients and for guiding decision making. The existing models, however, cannot be used during initial treatment or screening. This study aimed to derive and validate an echocardiography-based mortality prediction model for HD using deep learning (DL).
Methods:
In this multicenter retrospective cohort study, the subjects were admitted adult (age ≥ 18 years) HD patients who underwent echocardiography. The outcome was in-hospital mortality. We extracted predictor variables from echocardiography reports using text mining. We developed deep learning-based prediction model using derivation data of a hospital A. And we conducted external validation using echocardiography report of hospital B. We conducted subgroup analysis of coronary heart disease (CHD) and heart failure (HF) patients of hospital B and compared DL with the currently used predictive models (eg, Global Registry of Acute Coronary Events (GRACE) score, Thrombolysis in Myocardial Infarction score (TIMI), Meta-Analysis Global Group in Chronic Heart Failure (MAGGIC) score, and Get With The Guidelines-Heart Failure (GWTG-HF) score).
Results:
The study subjects comprised 25 776 patients with 1026 mortalities. The areas under the receiver operating characteristic curve (AUROC) of the DL model were 0.912, 0.898, 0.958, and 0.913 for internal validation, external validation, CHD, and HF, respectively, and these results significantly outperformed other comparison models.
Conclusions:
This echocardiography-based deep learning model predicted in-hospital mortality among HD patients more accurately than existing prediction models and other machine learning models.
More Related Videos
09:47Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
Published on: December 15, 2023
05:16Cutoff Value of Phase Angle by Bioelectrical Impedance Analysis at Admission as a Prognostic Factor in Patients with Acute Heart Failure
Published on: June 10, 2025
Related Concept Videos
00:59Hospitals-II
Nurses that work in...
01:28Hospitals-I
01:23Rheumatic Heart Disease I: Introduction
01:17Ischemic Heart Disease: Overview
Atherosclerosis, the primary malefactor, orchestrates this dangerous condition. It manifests as the accumulation of fatty deposits, akin to insidious plaques, within arterial walls. As time elapses, these plaques metamorphose, hardening and...
01:21Rheumatic Heart Disease III: Medical Management
02:27Predicting Molecular Geometry