Related Experiment Video
Updated: Sep 18, 2025
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
Published on: December 15, 2023
Heart sound classification based on convolutional neural network with convolutional block attention module
Ximing Huai1, Lei Jiang2,3, Chao Wang1
1Ningbo Key Laboratory of Intelligent Manufacturing of Textiles and Garments, Zhejiang Fashion Institute of Technology, Ningbo, Zhejiang, China.
Insights
This study enhances heart sound classification for diagnosing cardiovascular diseases (CVDs) using a novel attention-based Convolutional Neural Network (CNN). The improved model achieves high accuracy, showing promise for clinical applications.
Area of Science:
- Biomedical Engineering
- Artificial Intelligence in Medicine
- Cardiology
Background:
- Cardiovascular diseases (CVDs) are a primary cause of global mortality.
- Accurate and efficient diagnostic tools for CVDs are crucial.
- Current diagnostic methods may lack the precision needed for early detection.
Purpose of the Study:
- To develop an enhanced heart sound classification framework using deep learning.
- To integrate the Convolutional Block Attention Module (CBAM) with a Convolutional Neural Network (CNN).
- To evaluate the performance of the attention-based CNN for classifying heart sounds.
Main Methods:
- Utilized heart sound recordings from the PhysioNet CinC 2016 dataset.
- Processed audio data into spectrograms for analysis.
- Systematically evaluated twelve CNN models with varying CBAM configurations.
Main Results:
- The optimal CNN model with CBAM integration achieved 98.66% accuracy on the primary dataset.
- Validation on an independent PhysioNet 2022 dataset yielded 95.6% accuracy and 96.29% AUC.
- T-SNE visualizations demonstrated clear class separation, indicating effective feature extraction.
Conclusions:
- Selective integration of CBAM significantly improves CNN performance in heart sound classification.
- Attention-based architectures are effective for medical signal classification.
- The developed framework shows potential for real-world clinical application in diagnosing CVDs.
Abstract:
Cardiovascular diseases (CVDs) remain a leading cause of global mortality, underscoring the need for accurate and efficient diagnostic tools. This study presents an enhanced heart sound classification framework based on a Convolutional Neural Network (CNN) integrated with the Convolutional Block Attention Module (CBAM). Heart sound recordings from the PhysioNet CinC 2016 dataset were segmented and transformed into spectrograms, and twelve CNN models with varying CBAM configurations were systematically evaluated. Experimental results demonstrate that selectively integrating CBAM into early and mid-level convolutional blocks significantly improves classification performance. The optimal model, with CBAM applied after Conv Blocks 1-1, 1-2, and 2-1, achieved an accuracy of 98.66%, outperforming existing state-of-the-art methods. Additional validation using an independent test set from the PhysioNet 2022 database confirmed the model's generalization capability, achieving an accuracy of 95.6% and an AUC of 96.29%. Furthermore, T-SNE visualizations revealed clear class separation, highlighting the model's ability to extract highly discriminative features. These findings confirm the efficacy of attention-based architectures in medical signal classification and support their potential for real-world clinical applications.
Related Concept Videos
01:15Heart Sounds
Auscultation is the process of listening to these internal body sounds using a stethoscope. The heart produces four types of sounds, but only two—S1 and S2—can usually be heard with a stethoscope.
S1, also known as the "lub" sound, is caused by the closure of atrioventricular (A-V)...
01:30Classification of Signals
A continuous-time signal holds a value at every instant in time, representing information seamlessly. In contrast, a discrete-time signal holds values only at specific moments, often denoted as x(n), where...
01:19Conduction System of the Heart
The pacemaker cells are located in two primary nodes: the sinoatrial (SA) node and the atrioventricular (AV) node. The SA node pacemaker cells can autonomously depolarize, triggering an action potential that leads to the...
01:25Assessment of the Cardiovascular System IV: Auscultation
Normal Heart Sounds
S1 (First Heart Sound)-
S1 is made by the closure of the mitral and tricuspid valves (atrioventricular valves), marking the beginning of systole.
S2 (Second Heart Sound)-
S2 is made by the closure of the aortic and pulmonic valves (semilunar valves), marking the end of the systole.
01:30Heart Failure IV: Classification and Diagnostic Evaluation
01:22Force Classification
Contact and non-contact forces are two of the most widely used categories of forces. As the name suggests, contact forces require physical contact between two objects to act upon each other. Examples of contact forces include frictional,...