Screening for severe coronary stenosis in patients with apparently normal electrocardiograms based on deep learning
Zhengkai Xue1, Shijia Geng2, Shaohua Guo3
1Tianjin Key Laboratory of Ionic-Molecular Function of Cardiovascular Disease, Department of Cardiology, Tianjin Institute of Cardiology, The Second Hospital of Tianjin Medical University, Tianjin, China.
Insights
Deep learning models can identify severe coronary artery stenosis in patients with normal ECGs. Transfer learning combined with clinical data offers the most effective approach for early detection and treatment.
Area of Science:
- Cardiology
- Artificial Intelligence
- Medical Imaging
Background:
- Severe coronary artery stenosis can be difficult to detect in patients with normal electrocardiograms (ECGs).
- This diagnostic challenge can lead to missed treatment opportunities for at-risk individuals.
- Routine screenings may fail to identify critical cardiovascular conditions.
Purpose of the Study:
- To develop an effective deep learning (DL) model for distinguishing severe coronary stenosis from mild or no stenosis in patients presenting with normal ECGs.
- To evaluate the performance of DL models using ECG data alone versus combined with clinical information.
- To leverage transfer learning for improved feature extraction from ECG data.
Main Methods:
- Trained deep learning (DL) models from scratch and via transfer learning using ECG data from 392 patients (138 with severe stenosis).
- Evaluated models using ECG data solely and in conjunction with clinical factors (age, sex, hypertension, diabetes, dyslipidemia, smoking).
- Compared DL model performance against logistic regression using clinical data.
Main Results:
- DL models trained on ECG data alone showed limited sensitivity (54.5%) but good specificity (74.6%).
- Incorporating clinical data improved sensitivity (90.9%) but decreased specificity (42.3%) for DL models trained from scratch.
- The optimal model combined clinical data with an ECG transfer learning model, achieving an AUC of 0.847, 84.8% sensitivity, and 70.4% specificity.
Conclusions:
- Deep learning models effectively identify severe coronary stenosis in patients with normal ECGs.
- Transfer learning enhances ECG analysis by extracting 'deep features' efficiently.
- This approach offers a promising method for early detection and intervention in cardiovascular disease.
Background:
Patients with severe coronary arterystenosis may present with apparently normal electrocardiograms (ECGs), making it difficult to detect adverse health conditions during routine screenings or physical examinations. Consequently, these patients might miss the optimal window for treatment.
Methods:
We aimed to develop an effective model to distinguish severe coronary stenosis from no or mild coronary stenosis in patients with apparently normal ECGs. A total of 392 patients, including 138 with severe stenosis, were selected for the study. Deep learning (DL) models were trained from scratch and using pre-trained parameters via transfer learning. These models were evaluated based on ECG data alone and in combination with clinical information, including age, sex, hypertension, diabetes, dyslipidemia and smoking status.
Results:
We found that DL models trained from scratch using ECG data alone achieved a specificity of 74.6% but exhibited low sensitivity (54.5%), comparable to the performance of logistic regression using clinical data. Adding clinical information to the ECG DL model trained from scratch improved sensitivity (90.9%) but reduced specificity (42.3%). The best performance was achieved by combining clinical information with the ECG transfer learning model, resulting in an area under the receiver operating characteristic curve (AUC) of 0.847, with 84.8% sensitivity and 70.4% specificity.
Conclusions:
The findings demonstrate the effectiveness of DL models in identifying severe coronary stenosis in patients with apparently normal ECGs and validate an efficient approach utilizing existing ECG models. By employing transfer learning techniques, we can extract "deep features" that summarize the inherent information of ECGs with relatively low computational expense.
More Related Videos
Related Concept Videos
01:25Imaging Studies for Cardiovascular System VI: Calcium -Scoring CT
01:28Imaging Studies for Cardiovascular System V: CT
01:29Electrocardiogram
Three major waveforms are present in a typical ECG recording: the P wave, the QRS complex, and...
01:26Exercise Stress Test
Exercise stress testing, commonly known as a treadmill test, is a noninvasive procedure used to evaluate cardiovascular function and diagnose heart conditions.
Definition
An exercise stress test measures the heart's response to exertion using a treadmill or stationary bicycle. Chest electrodes record the heart's electrical activity through an ECG, and blood pressure is monitored regularly.
Purposes
01:17Imaging Studies for Cardiovascular System I:Echocardiography
Indications: Echocardiography is utilized to diagnose heart failure, valve disorders, and myocardial infarction. It also assesses cardiac structures' size, shape, and motion,...
01:20Imaging Studies for Cardiovascular System II:Types of Echocardiography
Types of Echocardiography
Transthoracic Echocardiography (TTE)
TTE is the most common type of echocardiogram which involves placing a transducer on the patient's chest, emitting sound waves to create heart images. TTE is invaluable for evaluating the heart's size, structure, and motion, making it particularly useful for...