Improving myocardial infarction diagnosis with Siamese network-based ECG analysis
Vaibhav Gadag1, Simrat Singh1, Anshul Harish Khatri1
1School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, Tamil Nadu, India.
Insights
This study presents a 98% accurate Siamese Network model for early detection of myocardial infarction (MI) using electrocardiography (ECG) images. The model effectively diagnoses MI, abnormal heartbeats, and normal rhythms, improving cardiovascular disease detection.
Area of Science:
- Cardiology
- Medical Imaging
- Machine Learning
Background:
- Myocardial infarction (MI) causes heart muscle damage due to poor blood flow and is a leading global cause of death.
- Diagnosing MI is challenging due to its asymptomatic nature and reliance on subjective electrocardiography (ECG) interpretation, prone to observer bias.
- Automated systems for MI detection from ECG images require high dependability, simplicity, and decision-making accuracy for clinical utility.
Purpose of the Study:
- To develop a dependable and accurate automated system for early detection of cardiovascular diseases using ECG images.
- To classify ECG images into three categories: Myocardial Infarction, Abnormal heartbeat, and Normal.
- To address the need for improved diagnostic tools in clinical settings for MI detection.
Main Methods:
- A dataset of 928 ECG images from Mendeley Data was utilized.
- ECG images were pre-processed and split into training (70%), validation (20%), and testing (10%) sets.
- A Siamese Network model was employed for training and classification.
Main Results:
- The developed model achieved a classification accuracy of 98%.
- The algorithm demonstrated excellent performance on imbalanced datasets by processing image pairs.
- Evaluation metrics for validation and testing datasets indicated near-perfect performance.
Conclusions:
- A Siamese Network model was successfully developed for early cardiovascular disease detection based on ECG signals.
- The study highlights the potential of automated ECG analysis for improved MI diagnosis.
- The model's high accuracy and ability to handle class imbalance offer a promising tool for clinical application.
Background:
Heart muscle damage from myocardial infarction (MI) is brought on by insufficient blood flow. The leading cause of death for middle-aged and older people worldwide is myocardial infarction (MI), which is difficult to diagnose because it has no symptoms. Clinicians must evaluate electrocardiography (ECG) signals to diagnose MI, which is difficult and prone to observer bias. To be effective in actual practice, an automated, and computerized detection system for Myocardial Infarction using ECG images, must meet a number of criteria.
Objective:
In an actual clinical situation, these requirements-such as dependability, simplicity, and superior decision-making abilities-remain crucial. In the current work, we have developed a model using a dataset that consists of a combination of 928 ECG images taken from publicly available Mendeley Data. It was converted into three classes Myocardial Infarction, Abnormal heartbeat, and Normal.
Methods:
The dataset is then imported, pre-processed, and split into a 70:20:10 ratio of training, validation, and testing. It is then trained using the Siamese Network Model.
Results:
The classification accuracy comes out to be 98%. The algorithm works excellently with datasets having class imbalance by taking pair of images as input. The validation and testing classification matrix is then generated and the evaluation metrics for both of them come out to be a near-perfect value.
Conclusion:
In this study, we developed the ECG signals based early detection of cardiovascular diseases with Siamese network model.
