EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Hierarchical Ensemble Learning with Feature-Specific Classifiers for Heart Sound Classification

Aruna P. Kharat, Vipulsangram K. Kadam

International Journal of Experimental Research and Review · 30.04.2026 · 50 · 110-118

DOI: 10.52756/ijerr.2026.v50.009 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Experimental Research and Review, online erschienen am 30.04.2026. Verfasst von Aruna P. Kharat und 1 weiteren Autor:innen. Die Arbeit nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Heart sound is one of the primary and most significant screening cardiac function screening techniques because it is non-invasive and requires fewer resources, especially in health facilities with limited resources. Nonetheless, manual evaluation is subjective and relies on clinicians' skills, potentially leading to variability and misdiagnosis. The proposal in this study is a machine learning ensemble design that will help automate the classification of phonocardiogram (PCG) signals into normal and abnormal categories. The proposed system combines time-domain, frequency-domain and time–frequency characteristics to store complete heart-sound characteristics. Three classifiers, random forest, Support Vector Machine (SVM) and gradient boosting, are used as base learners and aggregated using weighted voting to improve classification performance. Pre-processing, such as filtering, normalization and segmentation, is performed to improve signal quality and ensure accurate feature extraction. Experimental data show that the proposed model achieves 97.5% accuracy, 97.9% sensitivity and 96.7% specificity, outperforms more traditional single classifiers, and is equally effective as more complex deep learning models while incurring a lower computational cost and offering higher interpretability. The offered method is economical, dependable, and effective in the early detection of heart defects and, as such, is most applicable in primary care, rural settings and mobile diagnostic tools.

Zitieren

APA

Kharat, A. P., & Kadam, V. K. (2026). Hierarchical Ensemble Learning with Feature-Specific Classifiers for Heart Sound Classification. International Journal of Experimental Research and Review, 50, 110-118. https://doi.org/10.52756/ijerr.2026.v50.009

Vancouver

Kharat AP, Kadam VK. Hierarchical Ensemble Learning with Feature-Specific Classifiers for Heart Sound Classification. International Journal of Experimental Research and Review. 2026;50:110-118. doi:10.52756/ijerr.2026.v50.009

BibTeX
@article{Kharat2026,
  author = {Kharat, Aruna P. and Kadam, Vipulsangram K.},
  title = {Hierarchical Ensemble Learning with Feature-Specific Classifiers for Heart Sound Classification},
  journal = {International Journal of Experimental Research and Review},
  year = {2026},
  volume = {50},
  pages = {110--118},
  doi = {10.52756/ijerr.2026.v50.009}
}
Kostenlos starten →