Related Experiment Video
Updated: Sep 14, 2025
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
Published on: December 15, 2023
Hybrid deep-CNN and Bi-LSTM model with attention mechanism for enhanced ECG-based heart disease diagnosis
Gaurav Kumar1, Neeraj Varshney2
1Department of Computer Engineering and Applications, GLA University, Uttar Pradesh, Mathura, India.
Insights
This study introduces a novel Deep-CNN and Bi-LSTM model for accurate cardiovascular disease detection using electrocardiogram (ECG) data. The advanced model significantly improves diagnostic accuracy, addressing limitations of traditional methods.
Area of Science:
- Cardiology
- Biomedical Engineering
- Artificial Intelligence in Medicine
Background:
- Cardiovascular diseases (CVDs) are a leading cause of global mortality, with 17.9 million deaths annually.
- Electrocardiogram (ECG) analysis for CVD detection faces challenges due to signal variability and reliance on manual interpretation, leading to diagnostic errors.
- There is a critical need for automated, accurate, and reliable systems for early heart disease detection.
Purpose of the Study:
- To develop and evaluate an advanced deep learning model for enhanced cardiovascular disease diagnosis using ECG data.
- To improve the accuracy and reliability of heart disease classification by integrating Deep-Convolutional Neural Network (Deep-CNN) and Bidirectional Long Short-Term Memory (Bi-LSTM) with an Attention Mechanism.
- To overcome the limitations of conventional diagnostic methods and human analysis in ECG interpretation.
Main Methods:
- Utilized a hybrid deep learning architecture combining Deep-CNN for spatial feature extraction and Bi-LSTM for temporal dependency analysis.
- Incorporated an Attention Mechanism to focus on relevant ECG signal segments, improving diagnostic performance.
- Evaluated the model on the UCI Cleveland Heart Disease dataset, comprising 303 patient records and 14 clinical features.
Main Results:
- The proposed Deep-CNN and Bi-LSTM model achieved a high accuracy of 97.23%.
- The model demonstrated strong performance with a recall of 97.72%, precision, and an F1 score of 96.90%.
- The integrated approach outperformed existing boosting ensemble and hybrid models in diagnostic performance.
Conclusions:
- The developed hybrid deep learning model offers a significant advancement in automated cardiovascular disease detection from ECG data.
- The combination of Deep-CNN, Bi-LSTM, and Attention Mechanism effectively addresses ECG signal variability and enhances diagnostic accuracy.
- This approach holds promise for improving patient outcomes through earlier and more reliable heart disease diagnosis.
Abstract:
According to the World Health Organization (WHO), 17.9 million people die yearly from cardiovascular Diseases (CVDs), including heart attacks. Cardiovascular diseases, including heart attack, kill 32% of people globally. Current approaches struggle with electrocardiogram (ECG) signal variability, causing diagnosing errors. The adoption of automated and accurate models for heart disease detection is lacking since conventional methods rely on human analysis, which is time-consuming and error-prone. This work covers the crucial topic of heart disease diagnosis, especially ECG data analysis for cardiovascular disease detection. The integration of the Deep-Convolutional Neural Network (Deep-CNN) and Bidirectional Long Short-Term Memory (Bi-LSTM) model with an Attention Mechanism enhances the accuracy and reliability of heart disease categorisation. The Deep-CNN component efficiently extracts features from capture spatial linkages, while the Bi-LSTM layers handle temporal dependencies to identify patient health patterns over time. The model is evaluated on 303 patient records with 14 clinical characteristics from the University of California, Irvine (UCI) Cleveland Heart Disease dataset. The suggested technique has 97.23% accuracy, 97.72% recall, precision, and 96.90% F1 score. These findings show that the proposed architecture improves diagnostic performance more than boosting ensemble approaches and hybrid models.
