EUVIMEDEUVIMED entdecken
Open Access · CC BY-SA 4.0

BackMix-Enhanced Semi-Supervised Learning for Automated Detection of Aortic Stenosis from Transthoracic Echocardiographic Images

Fatima Ezzahra Elkouahy, Badreddine Labakoum, Hajar Ouahid, Mansoor Alam, Hamid El Malali, Lhoucine Ben Taleb, Azeddine Mouhsen

Journal of Electronics, Electromedical Engineering, and Medical Informatics · 18.07.2026 · 8 (3) · 1204-1223

DOI: 10.35882/jeeemi.v8i3.1657 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Electronics, Electromedical Engineering, and Medical Informatics, online erschienen am 18.07.2026. Verfasst von Fatima Ezzahra Elkouahy und 6 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Aortic stenosis (AS) is a prevalent and progressive valvular heart disease requiring accurate severity assessment for optimal clinical decision-making. Transthoracic echocardiography (TTE) is the standard diagnostic modality; however, its interpretation remains operator-dependent and subject to inter-observer variability. In this study, we propose an anatomically guided BackMix-enhanced semi-supervised learning framework for automated echocardiographic view classification and AS severity assessment. The approach leverages Gradient-weighted Class Activation Mapping (Grad-CAM) to preserve diagnostically relevant anatomical regions during augmentation while modifying background areas to mitigate shortcut learning. A semi-supervised self-training strategy combined with an ensemble classification framework was used to exploit both labeled and unlabeled data. The framework was evaluated on the TMED2 dataset of echocardiography, comprising 24,964 TTE images. To assess generalizability, external validation was performed on an independent dataset of 300 cardiologist-labeled echocardiographic images, including apical four-chamber (A4C), parasternal long-axis (PLAX), and parasternal short-axis (PSAX) views. Experimental results demonstrated that the proposed method outperformed the baseline semi-supervised model, improving view classification accuracy by approximately 3% and AS severity classification accuracy by 4–6%, with the largest gain observed in moderate AS. Performance remained consistent on the external validation dataset, supporting the robustness of the proposed approach. Statistical analysis confirmed the significance of these improvements (p < 0.01). Grad-CAM evaluation further demonstrated improved localization of clinically relevant regions. These findings suggest that anatomically guided BackMix augmentation combined with semi-supervised ensemble learning can improve classification accuracy, robustness, and interpretability in echocardiographic analysis under limited annotation conditions, offering a promising approach for automated AS assessment across independent clinical datasets

Zitieren

APA

Elkouahy, F. E., Labakoum, B., Ouahid, H., Alam, M., Malali, H. E., Taleb, L. B., & Mouhsen, A. (2026). BackMix-Enhanced Semi-Supervised Learning for Automated Detection of Aortic Stenosis from Transthoracic Echocardiographic Images. Journal of Electronics, Electromedical Engineering, and Medical Informatics, 8(3), 1204-1223. https://doi.org/10.35882/jeeemi.v8i3.1657

Vancouver

Elkouahy FE, Labakoum B, Ouahid H, Alam M, Malali HE, Taleb LB, et al. BackMix-Enhanced Semi-Supervised Learning for Automated Detection of Aortic Stenosis from Transthoracic Echocardiographic Images. Journal of Electronics, Electromedical Engineering, and Medical Informatics. 2026;8(3):1204-1223. doi:10.35882/jeeemi.v8i3.1657

BibTeX
@article{Elkouahy2026,
  author = {Elkouahy, Fatima Ezzahra and Labakoum, Badreddine and Ouahid, Hajar and Alam, Mansoor and Malali, Hamid El and Taleb, Lhoucine Ben and Mouhsen, Azeddine},
  title = {BackMix-Enhanced Semi-Supervised Learning for Automated Detection of Aortic Stenosis from Transthoracic Echocardiographic Images},
  journal = {Journal of Electronics, Electromedical Engineering, and Medical Informatics},
  year = {2026},
  volume = {8},
  number = {3},
  pages = {1204--1223},
  doi = {10.35882/jeeemi.v8i3.1657}
}
Kostenlos starten →