EUVIMEDEUVIMED entdecken
Crossref

Anatomical Pretext Tasks with Hybrid CNN-ViT Backbone for Enhanced SVM-Based Mammogram Analysis

Abdullah Ghanim Jaber, Abeer Ahmed Ali, Ali A. Mahmood, Mohammed Jamal Salim, Ghaith Jaafar Mohammed

Dijlah Journal of Engineering Sciences ISSN: 3078-9664, e-ISSN: 3078-9656 · 22.03.2026 · 3 (1)

DOI: 10.65204/djes.v3i1.652 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Dijlah Journal of Engineering Sciences ISSN: 3078-9664, e-ISSN: 3078-9656, online erschienen am 22.03.2026. Verfasst von Abdullah Ghanim Jaber und 4 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We suggest a new method of enhancing the feature discriminability of mammogram analysis through integrating anatomical pretext tasks in a hybrid CNN-ViT backbone followed by improved classification using SVM. Traditional self-supervised approaches usually use generic image transforms, which does not necessarily reflect the subtle clinical reasoning regarding radiologists. To tackle this, we formulate domain-specific pretext tasks, which directly model anatomical priors, such as spatial context reconstruction, orientation prediction conditional on ductal tree alignment and lesion-context consistency using contrastive learning. The hybrid backbone is a hybrid of ResNet-50 and ViT, which extracts local and global context patterns, respectively, to produce an integrated feature representation that links hierarchical and long-range dependencies. These characteristics are then trimmed down using PCA, so that they can be compatible with SVM kernels whilst retaining anatomical significance. In comparison to the current approaches, our framework is the only one to use the pretraining goals with clinical workflows, thus enhancing the interpretability of features and minimizing the need of large labeled datasets. Experiments indicate that the suggested approach performs better than the conventional deep feature extractors in mammogram classification tasks. Integrating domain-sensitive medical image analysis with self-directed learning through incorporating anatomical reasoning provides new opportunities to analyze medical images, especially in cases when few annotations are available. The contribution of the paper is the progression of the synergy between radiologist-inspired feature learning and up-to-date deep architectures, which gives a scalable algorithm that enhances the accuracy of diagnostic outcomes.

Zitieren

APA

Jaber, A. G., Ali, A. A., Mahmood, A. A., Salim, M. J., & Mohammed, G. J. (2026). Anatomical Pretext Tasks with Hybrid CNN-ViT Backbone for Enhanced SVM-Based Mammogram Analysis. Dijlah Journal of Engineering Sciences ISSN: 3078-9664, e-ISSN: 3078-9656, 3(1). https://doi.org/10.65204/djes.v3i1.652

Vancouver

Jaber AG, Ali AA, Mahmood AA, Salim MJ, Mohammed GJ. Anatomical Pretext Tasks with Hybrid CNN-ViT Backbone for Enhanced SVM-Based Mammogram Analysis. Dijlah Journal of Engineering Sciences ISSN: 3078-9664, e-ISSN: 3078-9656. 2026;3(1). doi:10.65204/djes.v3i1.652

BibTeX
@article{Jaber2026,
  author = {Jaber, Abdullah Ghanim and Ali, Abeer Ahmed and Mahmood, Ali A. and Salim, Mohammed Jamal and Mohammed, Ghaith Jaafar},
  title = {Anatomical Pretext Tasks with Hybrid CNN-ViT Backbone for Enhanced SVM-Based Mammogram Analysis},
  journal = {Dijlah Journal of Engineering Sciences ISSN: 3078-9664, e-ISSN: 3078-9656},
  year = {2026},
  volume = {3},
  number = {1},
  doi = {10.65204/djes.v3i1.652}
}
Kostenlos starten →