EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Benign Share Benefit to Malignant: Balanced Mixing on Feature Space for Imbalanced Breast Cancer Classification

Farchan Hakim Raswa iD, Muhammad Fadlurrohman, Bach-Tung Pham, Ika Candradewi iD, Afiahayati iD, Ming-Hsiang Su iD, Chung-I Huang iD, Kuo-Chen Li iD, Shih-Lun Chen iD, Yung-Hui Li iD, Jia-Ching Wang

Bioengineering · 30.06.2026 · 13 (7) · 765

DOI: 10.3390/bioengineering13070765 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioengineering, online erschienen am 30.06.2026. Verfasst von Farchan Hakim Raswa und 10 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science and Information Engineering, National Central University, Taoyuan 320317, Taiwan. Die Arbeit nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

A deep learning model with an imbalanced mammography dataset can bias models toward common benign BI-RADS categories and reduce recognition of less frequent malignant or high-risk categories. To address this issue, we propose B2M (Benign Share Benefit to Malignant), a model-agnostic framework for imbalance-aware multi-class BI-RADS classification in C-View mammography. B2M uses a two-phase training strategy that combines dual sampling with feature-space mixing. In Phase I, the model is trained with dual sampling, integrating instance-based and class-balanced sampling to increase minority-class representation while preserving majority-class diversity. In Phase II, the model is fine-tuned with feature-space mixing using samples from the two sampling streams. A soft-target regularization objective supervises the mixed features using labels from both streams, encouraging smoother decision boundaries across BI-RADS categories. We evaluated B2M on an imbalanced mammography cohort from the C-View EMBED dataset using stratified 5-fold cross-validation across multiple CNN backbones. C-View is a synthesized 2D mammographic image generated from 3D digital breast tomosynthesis data, capturing DBT-derived structural information while requiring less memory and computation than processing the full 3D image volume. Among these experiments, ResNeXt-50 with B2M achieved the highest balanced accuracy and Macro-F1 scores compared with the evaluated oversampling and mixing-based methods. This improvement requires an offline training-time overhead of approximately 2.81×, but it does not increase inference cost. Overall, the results suggest that B2M may be useful for imbalanced multi-class BI-RADS classification in C-View mammography. However, the findings are based on the EMBED cohort, and further validation, including external and prospective evaluation, is needed before clinical use.

Schlagwörter

Stichwörter der Autor:innen

breast cancer classificationclass imbalancedeep learningfeature-space augmentationmammography

Autor:innen und Institutionen

  • Department of Computer Science and Information Engineering, National Central University, Taoyuan 320317, Taiwan.
  • Department of Computer Science and Electronics, Universitas Gadjah Mada, Yogyakarta 55281, Indonesia.
  • Department of Data Science, Soochow University, Taipei 100006, Taiwan.
  • Department of Management Information Systems, National Chung Hsing University, Taichung 402202, Taiwan.
  • Department of Information Management, Chung Yuan Christian University, Taoyuan 320314, Taiwan.
  • Department of Electronic Engineering, Chung Yuan Christian University, Taoyuan 320314, Taiwan.

ORCID-Profile: Farchan Hakim Raswa, Ika Candradewi, Afiahayati, Ming-Hsiang Su, Chung-I Huang, Kuo-Chen Li, Shih-Lun Chen, Yung-Hui Li

Zitieren

APA

Raswa, F. H., Fadlurrohman, M., Pham, B. T., Candradewi, I., Afiahayati, Su, M. H., Huang, C. I., Li, K. C., Chen, S. L., Li, Y. H., & Wang, J. C. (2026). Benign Share Benefit to Malignant: Balanced Mixing on Feature Space for Imbalanced Breast Cancer Classification. Bioengineering, 13(7), 765. https://doi.org/10.3390/bioengineering13070765

Vancouver

Raswa FH, Fadlurrohman M, Pham BT, Candradewi I, Afiahayati, Su MH, et al. Benign Share Benefit to Malignant: Balanced Mixing on Feature Space for Imbalanced Breast Cancer Classification. Bioengineering. 2026;13(7):765. doi:10.3390/bioengineering13070765

BibTeX
@article{Raswa2026,
  author = {Raswa, Farchan Hakim and Fadlurrohman, Muhammad and Pham, Bach-Tung and Candradewi, Ika and Afiahayati and Su, Ming-Hsiang and Huang, Chung-I and Li, Kuo-Chen and Chen, Shih-Lun and Li, Yung-Hui and Wang, Jia-Ching},
  title = {Benign Share Benefit to Malignant: Balanced Mixing on Feature Space for Imbalanced Breast Cancer Classification},
  journal = {Bioengineering},
  year = {2026},
  volume = {13},
  number = {7},
  pages = {765},
  doi = {10.3390/bioengineering13070765}
}
Kostenlos starten →