EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Transformative Approaches in Breast Cancer Detection: Integrating Transformers into Computer-Aided Diagnosis for Histopathological Classification

Majed Alwateer iD, Amna Bamaqa, Mohamed Farsi, Mansourah Aljohani, Mohamed Shehata iD, Mostafa A Elhosseini iD

Bioengineering · 20.02.2025 · 12 (3) · 212

DOI: 10.3390/bioengineering12030212 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioengineering, online erschienen am 20.02.2025. Verfasst von Majed Alwateer und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, College of Computer Science and Engineering, Taibah University, Yanbu 46421, Saudi Arabia. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Breast cancer (BC) remains a leading cause of cancer-related mortality among women worldwide, necessitating advancements in diagnostic methodologies to improve early detection and treatment outcomes. This study proposes a novel twin-stream approach for histopathological image classification, utilizing both histopathologically inherited and vision-based features to enhance diagnostic precision. The first stream utilizes Virchow2, a deep learning model designed to extract high-level histopathological features, while the second stream employs Nomic, a vision-based transformer model, to capture spatial and contextual information. The fusion of these streams ensures a comprehensive feature representation, enabling the model to achieve state-of-the-art performance on the BACH dataset. Experimental results demonstrate the superiority of the twin-stream approach, with a mean accuracy of 98.60% and specificity of 99.07%, significantly outperforming single-stream methods and related studies. Statistical analyses, including paired t-tests, ANOVA, and correlation studies, confirm the robustness and reliability of the model. The proposed approach not only improves diagnostic accuracy but also offers a scalable and efficient solution for clinical applications, addressing the challenges of resource constraints and increasing diagnostic demands.

Schlagwörter

Stichwörter der Autor:innen

breast cancer (BC)deep learning (DL)digital histopathologytransformers

Autor:innen und Institutionen

  • Department of Computer Science, College of Computer Science and Engineering, Taibah University, Yanbu 46421, Saudi Arabia.
  • Department of Computer Science and Informatics, Applied College, Taibah University, Madinah 41461, Saudi Arabia.
  • Department of Information Systems, College of Computer Science and Engineering, Taibah University, Yanbu 46421, Saudi Arabia.
  • Department of Bioengineering, Speed School of Engineering, University of Louisville, Louisville, KY 40292, USA.
  • Computers and Control Systems Engineering Department, Faculty of Engineering, Mansoura University, Mansoura 35516, Egypt.

ORCID-Profile: Majed Alwateer, Mohamed Shehata, Mostafa A Elhosseini

Zitieren

APA

Alwateer, M., Bamaqa, A., Farsi, M., Aljohani, M., Shehata, M., & Elhosseini, M. A. (2025). Transformative Approaches in Breast Cancer Detection: Integrating Transformers into Computer-Aided Diagnosis for Histopathological Classification. Bioengineering, 12(3), 212. https://doi.org/10.3390/bioengineering12030212

Vancouver

Alwateer M, Bamaqa A, Farsi M, Aljohani M, Shehata M, Elhosseini MA. Transformative Approaches in Breast Cancer Detection: Integrating Transformers into Computer-Aided Diagnosis for Histopathological Classification. Bioengineering. 2025;12(3):212. doi:10.3390/bioengineering12030212

BibTeX
@article{Alwateer2025,
  author = {Alwateer, Majed and Bamaqa, Amna and Farsi, Mohamed and Aljohani, Mansourah and Shehata, Mohamed and Elhosseini, Mostafa A},
  title = {Transformative Approaches in Breast Cancer Detection: Integrating Transformers into Computer-Aided Diagnosis for Histopathological Classification},
  journal = {Bioengineering},
  year = {2025},
  volume = {12},
  number = {3},
  pages = {212},
  doi = {10.3390/bioengineering12030212}
}
Kostenlos starten →