Comparative evaluation of transformer and CNN architectures for breast cancer detection across histopathology, mammography, and ultrasound
Journal of Information & Optimization Sciences · 2026 · 47 (5-A) · 1613-1621
DOI: 10.47974/jios-2251 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Information & Optimization Sciences, 2026. Verfasst von Kalyani Ghuge und 1 weiteren Autor:innen.
Abstract (Originalsprache)
Automated analysis of images in various imaging modalities are beneficial in breast cancer screening and diagnosis. This paper compares four deep learning models Vision Transformer (ViT), EfficientNetV2 with CBAM, ConvMixer, and MobileNetV4 on three publicly available datasets of breast imaging (histopathology, mammography, ultrasound). We measure accuracy, recall, F1-score, computational cost, and suitability to modality. To underline the focus on the reproducibility and modality-specific analysis, the methodology underlines concise descriptions of every model in terms of mathematical intuition, architecture illustration, al-gorithmic steps. We have found that attention-augmented CNNs (CBAM-EfficientNetV2) perform well in terms of accuracy and robustness across modalities and MobileNetV4 has a high efficiency-accuracy trade-off to be deployed on resourceconstrained devices.
Zitieren
APA
Ghuge, K., & Saravanan, D. (2026). Comparative evaluation of transformer and CNN architectures for breast cancer detection across histopathology, mammography, and ultrasound. Journal of Information & Optimization Sciences, 47(5-A), 1613-1621. https://doi.org/10.47974/jios-2251
Vancouver
Ghuge K, Saravanan D. Comparative evaluation of transformer and CNN architectures for breast cancer detection across histopathology, mammography, and ultrasound. Journal of Information & Optimization Sciences. 2026;47(5-A):1613-1621. doi:10.47974/jios-2251
BibTeX
@article{Ghuge2026,
author = {Ghuge, Kalyani and Saravanan, D.},
title = {Comparative evaluation of transformer and CNN architectures for breast cancer detection across histopathology, mammography, and ultrasound},
journal = {Journal of Information & Optimization Sciences},
year = {2026},
volume = {47},
number = {5-A},
pages = {1613--1621},
doi = {10.47974/jios-2251}
}
EUVIMED