Related Experiment Video
Updated: Oct 16, 2025
Semi-automated Optical Heartbeat Analysis of Small Hearts
Published on: September 16, 2009
Cardiovascular Disease Recognition Based on Heartbeat Segmentation and Selection Process
Mehrez Boulares1,2, Reem Alotaibi1, Amal AlMansour1
1Information System Department, Computing College, King Abdulaziz University, Jeddah, Makkah 21589, Saudi Arabia.
Insights
This study introduces an artificial intelligence model using convolutional neural networks (CNNs) for accurate cardiovascular disease (CVD) detection from heart sounds. The AI model achieved high accuracy, demonstrating its potential for early CVD diagnosis.
Area of Science:
- Cardiology
- Biomedical Engineering
- Artificial Intelligence
Background:
- Cardiovascular disease (CVD) diagnosis often relies on cardiac auscultation, a method requiring expert interpretation of heart sounds (Phonocardiogram - PCG).
- Automated analysis of PCG signals using artificial intelligence (AI) offers a promising approach to aid physicians in preliminary CVD diagnosis.
Purpose of the Study:
- To develop an accurate cardiovascular disease recognition model utilizing unsupervised and supervised machine learning methods, specifically a convolutional neural network (CNN).
- To evaluate the performance of the proposed AI model on publicly available heart sound datasets (PASCAL and PhysioNet).
Main Methods:
- The study employed a convolutional neural network (CNN) architecture for analyzing Phonocardiogram (PCG) signals.
- Both unsupervised and supervised machine learning techniques were utilized within the CNN framework.
- The model's performance was rigorously evaluated on the PASCAL and PhysioNet heart sound datasets.
Main Results:
- Heart cycle segmentation and segment selection significantly impacted model performance metrics, including accuracy, sensitivity (TPR), precision (PPV), and specificity (TNR).
- On the PASCAL dataset, the model achieved an overall accuracy of 0.87, precision of 0.81, and sensitivity of 0.83.
- On the PhysioNet dataset, the model demonstrated superior performance with 0.97 accuracy, 0.946 sensitivity, 0.944 precision, and 0.946 specificity.
Conclusions:
- The developed AI model, based on CNNs, shows significant potential for accurate and reliable cardiovascular disease detection from heart sound analysis.
- The findings highlight the critical role of signal processing techniques like segmentation and segment selection in optimizing AI-driven diagnostic tools.
- The high performance on both PASCAL and PhysioNet datasets suggests the generalizability and clinical utility of the proposed approach for automated CVD screening.
Abstract:
Assessment of heart sounds which are generated by the beating heart and the resultant blood flow through it provides a valuable tool for cardiovascular disease (CVD) diagnostics. The cardiac auscultation using the classical stethoscope phonological cardiogram is known as the most famous exam method to detect heart anomalies. This exam requires a qualified cardiologist, who relies on the cardiac cycle vibration sound (heart muscle contractions and valves closure) to detect abnormalities in the heart during the pumping action. Phonocardiogram (PCG) signal represents the recording of sounds and murmurs resulting from the heart auscultation, typically with a stethoscope, as a part of medical diagnosis. For the sake of helping physicians in a clinical environment, a range of artificial intelligence methods was proposed to automatically analyze PCG signal to help in the preliminary diagnosis of different heart diseases. The aim of this research paper is providing an accurate CVD recognition model based on unsupervised and supervised machine learning methods relayed on convolutional neural network (CNN). The proposed approach is evaluated on heart sound signals from the well-known, publicly available PASCAL and PhysioNet datasets. Experimental results show that the heart cycle segmentation and segment selection processes have a direct impact on the validation accuracy, sensitivity (TPR), precision (PPV), and specificity (TNR). Based on PASCAL dataset, we obtained encouraging classification results with overall accuracy 0.87, overall precision 0.81, and overall sensitivity 0.83. Concerning Micro classification results, we obtained Micro accuracy 0.91, Micro sensitivity 0.83, Micro precision 0.84, and Micro specificity 0.92. Using PhysioNet dataset, we achieved very good results: 0.97 accuracy, 0.946 sensitivity, 0.944 precision, and 0.946 specificity.
Related Concept Videos
01:30Pulse rhythm
Conversely, an irregular pulse pattern is termed dysrhythmia, stemming from disruptions in cardiac...
01:25Cardiomyopathy I: Introduction and Classification
01:30Dysrhythmias V: Evaluating Dysrhythmias
01:29Electrocardiogram
Three major waveforms are present in a typical ECG recording: the P wave, the QRS complex, and...
01:29Disturbances in Heart Rhythm
Arrhythmias are categorized by their speed, rhythm, and origin. A slow heart...
01:13Special considerations while measuring pulse