April 11, 2026Scientific Reports3 citationsOpen Access

Advancing cardiovascular screening: deep learning-based heart-sound classification using SMOTE and temporal modeling

Q: What is the clinical evidence from this study?

Study design: Other. Population: Cardiac murmurs (n=8432). Intervention: Deep learning-based heart-sound classification framework (SMOTE + RNN). Primary outcome: Classification accuracy.

AAAsmaa Ameen IFIbrahim Eldesouky Fattoh TETarek Abd El‐Hafeez

Key Result

The proposed deep learning framework combining SMOTE and RNN achieved 98.6% accuracy on the PhysioNet 2022 dataset and 98.5% on the PhysioNet 2016 dataset for automated murmur classification.

Structured PICO

Does a deep learning framework using SMOTE and RNN accurately classify cardiac murmurs from phonocardiogram recordings?

Population

Phonocardiogram (PCG) recordings from the PhysioNet 2016 and PhysioNet 2022 datasets

Intervention

Deep learning-based heart-sound classification framework combining peak-based segmentation, Mel-Frequency Cepstral Coefficient (MFCC) feature extraction, Synthetic Minority Over-sampling Technique (SMOTE)-based class balancing, and Recurrent Neural Network (RNN)-driven temporal modeling

Outcome

Accuracy, precision, recall, and F1-score of murmur classificationsurrogate

A deep learning framework combining temporal modeling and balanced learning achieved high accuracy in automated cardiac murmur detection from phonocardiograms, highlighting its potential for scalable non-invasive screening.

Limitations

Data quality and generalizability (sensitivity to input quality, variations in recording devices/noise)
Synthetic oversampling via SMOTE may introduce overfitting risk
Computational complexity of the RNN architecture
Lack of interpretability (black box model)
Segmentation constraints (assumes regular cardiac cycles)

Abstract

Abstract Early and reliable detection of cardiac murmurs from phonocardiogram (PCG) recordings is essential for improving cardiovascular screening and supporting diagnosis in primary care. However, automated murmur classification remains challenging due to signal variability, class imbalance, and temporal dependence within heart-sound sequences. This study presents a leakage-safe heart-sound classification framework that combines peak-based segmentation, Mel-Frequency Cepstral Coefficient (MFCC) feature extraction, Synthetic Minority Over-sampling Technique (SMOTE)–based class balancing, and Recurrent Neural Network (RNN)–driven temporal modeling. Segmentation was performed around cardiac onset peaks, and evaluation was conducted using recording-level splits for the PhysioNet 2016 dataset and patient-level splits for the PhysioNet 2022 dataset to prevent segment correlation bias. The proposed model achieved 98.6% accuracy (precision = 98.26%, recall = 98.95%, F1-score = 98.61%) on PhysioNet 2022, and 98.5% accuracy (precision = 98.49%, recall = 98.52%, F1-score = 98.50%) on PhysioNet 2016, demonstrating consistently high performance across datasets with different class distributions. These results indicate that combining temporal modeling with balanced learning improves robustness in murmur detection. The findings highlight the potential of PCG-based deep learning systems to support scalable, non-invasive cardiac screening, particularly in settings with limited access to specialist assessment.

AI से पूछें

Bookmark

View Full Paper