EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Hierarchical Swin Transformer Ensemble with Explainable AI for Robust and Decentralized Breast Cancer Diagnosis

Md Redwan Ahmed iD, Hamdadur Rahman, Zishad Hossain Limon, Md Ismail Hossain Siddiqui, Mahbub Alam Khan, Al Shahriar Uddin Khondakar Pranta, Rezaul Haque iD, S M Masfequier Rahman Swapno iD, Young-Im Cho iD, Mohamed S Abdallah iD

Bioengineering · 13.06.2025 · 12 (6) · 651

DOI: 10.3390/bioengineering12060651 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioengineering, online erschienen am 13.06.2025. Verfasst von Md Redwan Ahmed und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science and Engineering, East West University, Dhaka 1212, Bangladesh. Die Arbeit wurde bisher 61-mal zitiert (Crossref) und nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Gachon University, Korea Agency for Technology and Standards.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Early and accurate detection of breast cancer is essential for reducing mortality rates and improving clinical outcomes. However, deep learning (DL) models used in healthcare face significant challenges, including concerns about data privacy, domain-specific overfitting, and limited interpretability. To address these issues, we propose BreastSwinFedNetX, a federated learning (FL)-enabled ensemble system that combines four hierarchical variants of the Swin Transformer (Tiny, Small, Base, and Large) with a Random Forest (RF) meta-learner. By utilizing FL, our approach ensures collaborative model training across decentralized and institution-specific datasets while preserving data locality and preventing raw patient data exposure. The model exhibits strong generalization and performs exceptionally well across five benchmark datasets—BreakHis, BUSI, INbreast, CBIS-DDSM, and a Combined dataset—achieving an F1 score of 99.34% on BreakHis, a PR AUC of 98.89% on INbreast, and a Matthews Correlation Coefficient (MCC) of 99.61% on the Combined dataset. To enhance transparency and clinical adoption, we incorporate explainable AI (XAI) through Grad-CAM, which highlights class-discriminative features. Additionally, we deploy the model in a real-time web application that supports uncertainty-aware predictions and clinician interaction and ensures compliance with GDPR and HIPAA through secure federated deployment. Extensive ablation studies and paired statistical analyses further confirm the significance and robustness of each architectural component. By integrating transformer-based architectures, secure collaborative training, and explainable outputs, BreastSwinFedNetX provides a scalable and trustworthy AI solution for real-world breast cancer diagnostics.

Schlagwörter

Stichwörter der Autor:innen

breast cancerclinical decision supportensemble learningfederated learningprivacy-preservingvision transformers

Autor:innen und Institutionen

  • Department of Computer Science and Engineering, East West University, Dhaka 1212, Bangladesh.
  • Department of Management Information System, International American University, 3440 Wilshire Blvd. STE 1000, Los Angeles, CA 90010, USA.
  • Department of Computer Science, Westcliff University, Irvine, CA 92614, USA.
  • Department of Engineering/Industrial Management, Westcliff University, Irvine, CA 92614, USA.
  • Department of Management Information System, Pacific State University, 3424 Wilshire Blvd., 12th Floor, Los Angeles, CA 90010, USA.
  • Department of Computer Science, Wright State University, 3640 Colonel Glenn Hwy, Dayton, OH 45435, USA.

ORCID-Profile: Md Redwan Ahmed, Rezaul Haque, S M Masfequier Rahman Swapno, Young-Im Cho, Mohamed S Abdallah

Förderung

  • Gachon University
  • Korea Agency for Technology and Standards

Zitieren

APA

Ahmed, M. R., Rahman, H., Limon, Z. H., Siddiqui, M. I. H., Khan, M. A., Pranta, A. S. U. K., Haque, R., Swapno, S. M. M. R., Cho, Y. I., & Abdallah, M. S. (2025). Hierarchical Swin Transformer Ensemble with Explainable AI for Robust and Decentralized Breast Cancer Diagnosis. Bioengineering, 12(6), 651. https://doi.org/10.3390/bioengineering12060651

Vancouver

Ahmed MR, Rahman H, Limon ZH, Siddiqui MIH, Khan MA, Pranta ASUK, et al. Hierarchical Swin Transformer Ensemble with Explainable AI for Robust and Decentralized Breast Cancer Diagnosis. Bioengineering. 2025;12(6):651. doi:10.3390/bioengineering12060651

BibTeX
@article{Ahmed2025,
  author = {Ahmed, Md Redwan and Rahman, Hamdadur and Limon, Zishad Hossain and Siddiqui, Md Ismail Hossain and Khan, Mahbub Alam and Pranta, Al Shahriar Uddin Khondakar and Haque, Rezaul and Swapno, S M Masfequier Rahman and Cho, Young-Im and Abdallah, Mohamed S},
  title = {Hierarchical Swin Transformer Ensemble with Explainable AI for Robust and Decentralized Breast Cancer Diagnosis},
  journal = {Bioengineering},
  year = {2025},
  volume = {12},
  number = {6},
  pages = {651},
  doi = {10.3390/bioengineering12060651}
}
Kostenlos starten →