Breast cancer detection improved by lightweight CNN transformer model
A Lightweight CNN Integrated Compact Convolutional Transformer for Multi-Scale Feature Learning and reducing computational complexity for breast cancer mammography image detection and classification
Machine LearningArtificial IntelligenceComputer Vision and Pattern Recognition
Summary
Detecting and classifying breast cancer from mammogram images can be improved by combining CNNs with transformer models. The authors created a small, efficient model that captures details far apart in an image, which CNNs alone struggle with. Their model works well on multiple breast cancer datasets with high accuracy and uses fewer resources. They also included explanations of the model’s decisions to help doctors trust its results.
What this means in practice
- •For clinical imaging teams: Implement more accurate and efficient breast cancer diagnosis tools that run well in resource-limited hospitals using the lightweight CNN-transformer model.
- •For medical device developers: Incorporate compact CNN-transformer architectures into mammography analysis software to improve detection accuracy without heavy computational costs.$Commercial implications: Enables creation of cost-effective breast cancer screening products suitable for clinical use with limited hardware.
Authors
Md Taimur Ahad, Ainuddin Ahmed
Abstract
Over the years, Convolutional Neural Networks (CNNs) have demonstrated strong capability in cancer detection and classification using medical images. However, CNN-based models often struggle to capture long-range contextual dependencies. In such scenarios, integrating Compact Convolutional Transformer (CCT) architectures after the CCT layer allows CNN-extracted features to reshape into compact patch tokens using a CCT tokenizer, followed by the addition of positional embeddings to preserve spatial structure. Using 5-fold cross-validation, the model was tested on 3 sets of breast cancer mammography. With only 250,435 parameters, the model achieved 99%-100% accuracy across 3 datasets, indicating robust generalization. Explainable AI (XAI) was integrated into the model to explain the breast cancer classification process to enhance clinical trust. The results indicate that the proposed framework is suitable for computer-aided diagnosis systems, particularly in resource-constrained clinical environments. The novelty of the proposed CNN-integrated CCT overcomes the limitation of CNN's gradient degradation in the last layers by integrating convolutional tokenization with transformer-based learning. Lighter than ViT, which is effective in capturing long-range dependencies, the model has also proven efficient in breast cancer classification by capturing long-range dependencies among breast tissue regions.