Enhanced Heart Disease Classification Using Dual Attention Mechanisms and 3D-Echo Fusion Algorithm in Echocardiogram Videos
Heart disease remains a leading cause of mortality worldwide, making early detection and diagnosis crucial for preventing severe outcomes. Echocardiogram based classification of heart valve conditions is vital for effectively diagnosing and managing heart disease. In this paper, we present a novel h...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10816430/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Heart disease remains a leading cause of mortality worldwide, making early detection and diagnosis crucial for preventing severe outcomes. Echocardiogram based classification of heart valve conditions is vital for effectively diagnosing and managing heart disease. In this paper, we present a novel hybrid deep learning framework that integrates convolutional neural networks (CNNs) with recurrent neural networks (RNNs) alongside a 3D-Echo Fusion approach and a Dual Attention Model for heart valve disease classification using echocardiogram videos. Our method employs a CNN-based encoder to extract spatial features from individual video frames, which are then enhanced through a 3D-Echo Fusion technique to capture comprehensive volumetric information. An RNN models the temporal dynamics of heart valve motion, while the Dual Attention Model refines the classification process by enabling the model to focus on the most relevant features. This integration results in a classification accuracy of 98.7% and an AUC score of 0.85, demonstrating the model’s effectiveness in heart valve disease classification. The high accuracy and interpretability of the model provide potential clinical benefits by supporting decision-making and improving patient care. Furthermore, our proposed method addresses challenges related to the variability and complexity of cardiac data, aligning with current research that advocates for automated learning models in interpreting echocardiographic imagery. By leveraging advanced deep learning techniques, including 3D-Echo Fusion, our approach promises high classification accuracy while reducing reliance on manual annotations, thus enhancing scalability in cardiac diagnostics. |
|---|---|
| ISSN: | 2169-3536 |