EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Heart Sound Classification for Early Detection of Cardiovascular Diseases Using XGBoost and Engineered Acoustic Features

P P Satya Karthikeya iD, P Rohith, B Karthikeya, M Karthik Reddy, Akhil V M iD, Andrea Tigrini iD, Agnese Sbrollini iD, Laura Burattini iD

Sensors · 2026

DOI: 10.3390/s26020630 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, 2026. Verfasst von P P Satya Karthikeya und 7 weiteren Autor:innen. Beteiligte Einrichtung: Amrita School of Artificial Intelligence, Coimbatore, Amrita Vishwa Vidyapeetham, Amrita University, Coimbatore 641112, India. Hauptthemen laut MEDLINE: Cardiovascular Diseases, Heart Sounds. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Heart sound-based detection of cardiovascular diseases is a critical task in clinical diagnostics, where early and accurate identification can significantly improve patient outcomes. In this study, we investigate the effectiveness of combining traditional acoustic features and transformer-based Wav2Vec embeddings with advanced machine learning models for multi-class classification of five heart sound categories. Ten engineered acoustic features, i.e., Log Mel, MFCC, delta, delta-delta, chroma, discrete wavelet transform, zero-crossing rate, energy, spectral centroid, and temporal flatness, were extracted as regular features. Four model configurations were evaluated: a hybrid CNN + LSTM and XGBoost trained with either regular features or Wav2Vec embeddings. Models were assessed using a held-out test set with hyperparameter tuning and cross-validation. Results demonstrate that models trained on regular features consistently outperform Wav2Vec-based models, with XGBoost achieving the highest accuracy of 99%, surpassing the hybrid model at 98%. These findings highlight the importance of domain-specific feature engineering and the effectiveness of ensemble learning with XGBoost for robust and accurate heart sound classification, offering a promising approach for early detection and intervention in cardiovascular diseases.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Cardiovascular DiseasesHeart SoundsHumansBoosting Machine Learning AlgorithmsAcousticsMachine LearningClassification AlgorithmsEarly DiagnosisSignal Processing, Computer-AssistedConvolutional Neural NetworksAlgorithms

Stichwörter der Autor:innen

cardiovascular disease detectiondeep learningheart sound classificationmachine learning

Autor:innen und Institutionen

  • Amrita School of Artificial Intelligence, Coimbatore, Amrita Vishwa Vidyapeetham, Amrita University, Coimbatore 641112, India.
  • Department of Information Engineering, Università Politecnica delle Marche, 60131 Ancona, Italy.

ORCID-Profile: P P Satya Karthikeya, Akhil V M, Andrea Tigrini, Agnese Sbrollini, Laura Burattini

Zitieren

APA

Karthikeya, P. P. S., Rohith, P., Karthikeya, B., Reddy, M. K., V M, A., Tigrini, A., Sbrollini, A., & Burattini, L. (2026). Heart Sound Classification for Early Detection of Cardiovascular Diseases Using XGBoost and Engineered Acoustic Features. Sensors. https://doi.org/10.3390/s26020630

Vancouver

Karthikeya PPS, Rohith P, Karthikeya B, Reddy MK, V M A, Tigrini A, et al. Heart Sound Classification for Early Detection of Cardiovascular Diseases Using XGBoost and Engineered Acoustic Features. Sensors. 2026. doi:10.3390/s26020630

BibTeX
@article{Karthikeya2026,
  author = {Karthikeya, P P Satya and Rohith, P and Karthikeya, B and Reddy, M Karthik and V M, Akhil and Tigrini, Andrea and Sbrollini, Agnese and Burattini, Laura},
  title = {Heart Sound Classification for Early Detection of Cardiovascular Diseases Using XGBoost and Engineered Acoustic Features},
  journal = {Sensors},
  year = {2026},
  doi = {10.3390/s26020630}
}
Kostenlos starten →