Related Experiment Video
Updated: May 7, 2026
04:23
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
Published on: April 21, 2023
2.6K
Transformer based breast cancer classification from histopathological images using sequential residual recurrent
V Rajeswari1, N Kathirvel2, C Kumar3
1Computer Science and Technology, Karpagam College of Engineering, Coimbatore, Tamil Nadu, 641032, India.
Scientific Reports
|May 5, 2026
Summary
A new deep learning model, BrCTransNet, accurately classifies breast cancer from histopathology images. This advanced transformer network achieves high accuracy, offering a robust tool for improved breast cancer diagnosis.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Oncology
Background:
- Breast cancer remains a significant global health concern for women.
- Accurate and reliable diagnostic systems are crucial for effective treatment.
- Histopathological image analysis is key for breast cancer classification.
Purpose of the Study:
- To introduce a novel deep learning framework, BrCTransNet, for breast cancer classification.
- To enhance diagnostic accuracy and robustness using advanced AI techniques.
- To develop a reliable tool for breast cancer diagnosis from histopathological images.
Main Methods:
- Image preprocessing using Adaptive Gaussian Bilateral (AGB) filter.
- Feature extraction via Sequential Residual Recurrent Multiscale Attention Network (S2RMANet).
- Classification using Optimized Polarized Self-Attention-based Transformer (OPATransNet) with Chebyshev StarFish Optimization (ChStF).
Main Results:
- BrCTransNet achieved 98.97% classification accuracy on the Breast Histopathology Images dataset.
- The model demonstrated high recall (98.91%), F1-score (98.98%), and specificity (98.36%).
- Outperformed existing models in accuracy, robustness, and convergence speed.
Conclusions:
- The proposed BrCTransNet is a highly accurate and robust deep learning model for breast cancer classification.
- The integration of AGB filter, S2RMANet, OPATransNet, and ChStF significantly improves diagnostic performance.
- BrCTransNet shows great potential as a reliable tool for clinical breast cancer diagnosis.