EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Cardio-Dense: Diagnosis of Cardiac Abnormalities Based on Phonocardiogram Using Improved Swin Transformer Through Lightweight Dense Blocks

Alaa E S Ahmed iD, Mostafa E A Ibrahim iD, Yassine Daadaa iD

Diagnostics · 07.05.2026 · 16 (10) · 1421

DOI: 10.3390/diagnostics16101421 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Diagnostics, online erschienen am 07.05.2026. Verfasst von Alaa E S Ahmed und 2 weiteren Autor:innen. Beteiligte Einrichtung: College of Computer and Information Sciences, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh 11432, Saudi Arabia. Die Arbeit nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: This work was supported and funded by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University (IMSIU) (grant number IMSIU-DDRSP2601)..

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Cardiovascular diseases (CVDs) are among the top sources of mortality worldwide. To properly diagnose cardiovascular diseases, a low-cost remedy based on phonocardiography (PCG) signals must be proposed. Several deep learning (DL)-driven CVD systems are now being developed to identify various phases of the disease. Nevertheless, the approaches’ accuracy falls short of expectations, and they necessitate substantial processing resources and training data.

Methods

This paper proposes Cardio-Dense, a hybrid framework for multi-class CVD detection from phonocardiogram signals. The PCG waveform is first denoised in the wavelet domain and then converted into a 2D time–frequency spectrogram using continuous wavelet transform (CWT). We design a joint architecture that combines a Swin transformer for capturing global contextual dependencies with lightweight DenseBlocks for efficient local feature refinement, enabling robust learning from PCG spectrograms across five disease classes.

Results

Experiments on PCG datasets achieve up to 0.977 accuracy, 0.975 sensitivity, 0.992 specificity, 0.978 F1-score, 0.978 AUC, and 0.976 precision, while maintaining low computational overhead suitable for real-time inference.

Conclusions

The findings indicate that the proposed model provides an economical, non-invasive method for preliminary signal-level identification of multi-class heart valve diseases. It benefits clinicians by decreasing the need for arduous and error-prone manual PCG analysis. Furthermore, it offers quick, near-real-time categorization suitable for clinical and portable applications.

Schlagwörter

Stichwörter der Autor:innen

DenseBlocks architectureSwin transformerscardiovascular disorderclassificationcontinuous wavelet transformdeep learningphonocardiography

Autor:innen und Institutionen

  • College of Computer and Information Sciences, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh 11432, Saudi Arabia.

ORCID-Profile: Alaa E S Ahmed, Mostafa E A Ibrahim, Yassine Daadaa

Förderung

  • This work was supported and funded by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University (IMSIU) (grant number IMSIU-DDRSP2601).

Zitieren

APA

Ahmed, A. E. S., Ibrahim, M. E. A., & Daadaa, Y. (2026). Cardio-Dense: Diagnosis of Cardiac Abnormalities Based on Phonocardiogram Using Improved Swin Transformer Through Lightweight Dense Blocks. Diagnostics, 16(10), 1421. https://doi.org/10.3390/diagnostics16101421

Vancouver

Ahmed AES, Ibrahim MEA, Daadaa Y. Cardio-Dense: Diagnosis of Cardiac Abnormalities Based on Phonocardiogram Using Improved Swin Transformer Through Lightweight Dense Blocks. Diagnostics. 2026;16(10):1421. doi:10.3390/diagnostics16101421

BibTeX
@article{Ahmed2026,
  author = {Ahmed, Alaa E S and Ibrahim, Mostafa E A and Daadaa, Yassine},
  title = {Cardio-Dense: Diagnosis of Cardiac Abnormalities Based on Phonocardiogram Using Improved Swin Transformer Through Lightweight Dense Blocks},
  journal = {Diagnostics},
  year = {2026},
  volume = {16},
  number = {10},
  pages = {1421},
  doi = {10.3390/diagnostics16101421}
}
Kostenlos starten →