Improved Breast Cancer Classification through Combining Transfer Learning and Attention Mechanism

Abstract

Breast cancer, a leading cause of female mortality worldwide, poses a significant health challenge. Recent advancements in deep learning techniques have revolutionized breast cancer pathology by enabling accurate image classification. Various imaging methods, such as mammography, CT, MRI, ultrasound, and biopsies, aid in breast cancer detection. Computer-assisted pathological image classification is of paramount importance for breast cancer diagnosis. This study introduces a novel approach to breast cancer histopathological image classification. It leverages modified pre-trained CNN models and attention mechanisms to enhance model interpretability and robustness, emphasizing localized features and enabling accurate discrimination of complex cases. Our method involves transfer learning with deep CNN models—Xception, VGG16, ResNet50, MobileNet, and DenseNet121—augmented with the convolutional block attention module (CBAM). The pre-trained models are finetuned, and the two CBAM models are incorporated at the end of the pre-trained models. The models are compared to state-of-the-art breast cancer diagnosis approaches and tested for accuracy, precision, recall, and F1 score. The confusion matrices are used to evaluate and visualize the results of the compared models. They help in assessing the models’ performance. The test accuracy rates for the attention mechanism (AM) using the Xception model on the “BreakHis” breast cancer dataset are encouraging at 99.2% and 99.5%. The test accuracy for DenseNet121 with AMs is 99.6%. The proposed approaches also performed better than previous approaches examined in the related studies.

Bridging Discoveries and Treatments: The New Landscape of Breast Cancer Research

“…The article “Improved Breast Cancer Classification through Combining Transfer Learning and Attention Mechanism” introduces a novel approach that enhances the accuracy and interpretability of breast cancer histopathological image classification [ 5 ]. This method utilizes modified pre-trained Convolutional Neural Network (CNN) models and attention mechanisms to emphasize localized features and enable accurate discrimination in complex cases.…”

“…This Special Issue embodies our collective quest to understand the complexities of breast cancer through cutting-edge research and to translate these discoveries into actionable treatments that improve patient outcomes. Through a multidisciplinary lens, we explore innovative diagnostic tools [ 5 , 10 ], breakthrough therapies [ 4 , 9 ], and pioneering surgical techniques [ 11 ] that are reshaping the way we approach this disease. Our contributors, leading experts in their fields, offer insights into the evolving paradigms of breast cancer management, from molecular genetics to personalized medicine.…”

“…Each paper contributes uniquely to the evolving landscape of breast cancer research, ranging from the clinicopathological characteristics of breast carcinoma with neuroendocrine features [ 1 ], the identification of hub genes in the context of non-alcoholic fatty liver disease and triple-negative breast cancer [ 2 ], to advancements in diagnostic techniques [ 3 ], and the exploration of immunotherapy biomarkers [ 4 ]. Some of the topics covered include innovative methods for breast cancer classification combining transfer learning and attention mechanisms [ 5 ] and the role of diffusion-weighted imaging in breast cancer diagnosis among young patients [ 6 ]. The collective insights presented here not only underscore the complexity of this disease but also highlight the promising pathways toward more effective treatments and improved patient outcomes.…”

Optimal deep transfer learning driven computer‐aided breast cancer classification using ultrasound images

“…In Ashurov et al (2023), a novel approach is introduced for breast cancer histopathological image classification. It leverages modified pretrained CNN models and attention mechanisms to enhance model interpretability and robustness.…”