Related Experiment Video
Updated: Jul 31, 2025
Assessing Cardiac Reprogramming using High Content Imaging Analysis
Published on: October 26, 2020
Identification of patients with dilated phase of hypertrophic cardiomyopathy using a convolutional neural network
Naomi Hirota1, Shinya Suzuki1, Jun Motogi2
1Department of Cardiovascular Medicine, The Cardiovascular Institute, Tokyo, Japan.
Insights
An artificial intelligence model can detect dilated hypertrophic cardiomyopathy (dHCM) using electrocardiography (ECG). A single V5 lead ECG shows comparable performance to an eight-lead ECG for dHCM screening.
Area of Science:
- Cardiology
- Artificial Intelligence
- Medical Diagnostics
Background:
- Hypertrophic cardiomyopathy (HCM) can progress to a dilated phase (dHCM).
- Early detection of dHCM is crucial for patient management.
- Digital electrocardiography (ECG) is a widely accessible diagnostic tool.
Purpose of the Study:
- To develop an AI-derived model for detecting dHCM from digital ECGs.
- To evaluate the model's performance using multi-lead and single-lead ECGs.
- To determine if single-lead ECG can serve as an alternative for dHCM screening.
Main Methods:
- Retrospective analysis of 17,378 digital ECGs from a prospective cohort.
- Development of a convolutional neural network (CNN) model for dHCM detection.
- Evaluation of the CNN model using eight-lead, double-lead, and single-lead ECGs.
Main Results:
- The CNN model achieved an AUC of 0.929 with eight-lead ECG for dHCM detection.
- The model demonstrated superior performance with a single V5 lead ECG, achieving an AUC of 0.953.
- A single V5 lead ECG provided performance comparable to eight-lead ECG.
Conclusions:
- An AI-derived model effectively detects dHCM using digital ECGs.
- A single V5 lead ECG is a viable alternative to multi-lead ECG for dHCM screening.
- This approach can enhance the accessibility and efficiency of dHCM detection.
Background:
This study sought to develop an artificial intelligence-derived model to detect the dilated phase of hypertrophic cardiomyopathy (dHCM) on digital electrocardiography (ECG) and to evaluate the performance of the model applied to multiple-lead or single-lead ECG.
Methods:
This is a retrospective analysis using a single-center prospective cohort study (Shinken Database 2010-2017, n = 19,170). After excluding those without a normal P wave on index ECG (n = 1,831) and adding dHCM patients registered before 2009 (n = 39), 17,378 digital ECGs were used. Totally 54 dHCM patients were identified of which 11 diagnosed at baseline, 4 developed during the time course, and 39 registered before 2009. The performance of the convolutional neural network (CNN) model for detecting dHCM was evaluated using eight-lead (I, II, and V1-6), single-lead, and double-lead (I, II) ECGs with the five-fold cross validation method.
Results:
The area under the curve (AUC) of the CNN model to detect dHCM (n = 54) with eight-lead ECG was 0.929 (standard deviation [SD]: 0.025) and the odds ratio was 38.64 (SD 9.10). Among the single-lead and double-lead ECGs, the AUC was highest with the single lead of V5 (0.953 [SD: 0.038]), with an odds ratio of 58.89 (SD:68.56).
Conclusion:
Compared with the performance of eight-lead ECG, the most similar performance was achieved with the model with a single V5 lead, suggesting that this single-lead ECG can be an alternative to eight-lead ECG for the screening of dHCM.
Related Concept Videos
01:29Cardiomyopathy III: Hypertrophic Cardiomyopathy
01:30Cardiomyopathy II: Dilated Cardiomyopathy
01:25Cardiomyopathy I: Introduction and Classification