Convolutional neural networks (CNN) have attracted much attention as a classification method for histopathological images. However, the convolution of a single CNN model relies only on first-order statistics, which cannot describe the global information, and the average pooling layer has a weak ability to express local features, ignoring the subtle complexity of capturing nonlinear deep features (second-order statistics). Therefore, this paper proposes a breast cancer pathology image classification model based on the Dense model with joint higher-order dual attention mechanism(HDAM)and covariance pooling(CP) layer: HDCP-Dense, in which the HDAM module implements the interaction between channel and spatial attention and captures the global complementary features of the image; the covariance pooling(CP)layer adopts matrix square root normalisation, which not only solves the defects of the average pooling layer, but also further enhances the higher-order information of the features.Comparison experiments on the BreaKHis dataset and ablation experiments illustrate that the classification accuracy of this paper’s model, HDCP-Dense, can reach up to 98.67%, effectively achieves the classification of breast cancer pathology images.
LUKASIEWICZS, CZECZELEWSKIM, FORMAA, et al. Breast cancer-epidemiology, risk factors, classification, prognostic markers, and current treatment strategies:an updated review[J]. Cancers, 2021, 13(17): 4287.
[2]
VYV P T, YAOM M, LEN Q K, et al. Machine learning algorithm for distinguishing ductal carcinoma in situ from invasive breast cancer[J]. Cancers, 2022, 14(10): 2437.
[3]
AVERYA, ALDENJ, KRAMISHC, et al. The pathologic diagnosis of carcinoma in various tissues[J]. Advances in Cancer Research, 2022, 154: 1-14.
[4]
PHDH S.Breast cancer statistics, 2022[J]. CA: A Cancer Journal for Clinicians, 2022, 72(6): 524-541.
[5]
AGGARWALK, MIJWILM M, AL-MISTAREHIA H, et al. Has the future started?The current growth of artificial intelligence, machine learning, and deep learning[J]. Iraqi Journal for Computer Science and Mathematics, 2022, 3(1): 115-123.
[6]
HUANGZ Y. Research on single image super-resolution based on very deep super-resolution convolutional neural network[J]. Journal of Measurement Science and Instrumentation, 2022, 13(2): 276-283.
[7]
OYELADEO N, EZUGWUA E, VENTERH S, et al. Abnormality classification and localization using dual-branch whole-region-based CNN model with histopathological images[J]. Computers in Biology and Medicine, 2022, 149: 105943.
[8]
XIAH, QINY, TANY, et al. BA-Net: Brightness prior guided attention network for colonic polyp segmentation[J]. Biocybernetics and Biomedical Engineering, 2023, 43(3): 603-615.
[9]
FENGY, ATABANSIC C, NIEJ, et al. Multi-stage fully convolutional network for precise prostate segmentation in ultrasound images[J]. Biocybernetics and Biomedical Engineering, 2023, 43(3): 586-602.
[10]
ALLAMJ P, SAMANTRAYS, SAHOOS P, et al. A deformable CNN architecture for predicting clinical acceptability of ECG signal[J]. Biocybernetics and Biomedical Engineering, 2023, 43(1): 335-351.
[11]
XUZ, LIM S, LUY, et al. Reversed domain adaptation for nuclei segmentation-based pathological image classification[J]. Computers in Biology and Medicine, 2024, 168: 107726.
[12]
HOUY, ZHANGW, CHENGR, et al. Meta-adaptive-weighting-based bilateral multi-dimensional refined space feature attention network for imbalanced breast cancer histopathological image classification[J]. Computers in Biology and Medicine, 2023, 164: 107300.
[13]
XUC, YIK, JIANGN, et al. MDFF-Net: a multi-dimensional feature fusion network for breast histopathology image classification[J]. Computers in Biology and Medicine, 2023, 165: 107385.
[14]
LUS Y, WANGS H, ZHANGY D. BCDNet: an optimized deep network for ultrasound breast cancer detection[J]. IRBM, 2023, 44(4): 100774.
[15]
HUANGG, LIUZ, VAN DER MAATENL, et al. Densely connected convolutional networks[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017: 2261-2269.
[16]
SZEGEDYC, LIUW, JIAY, et al. Going deeper with convolutions[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015: 1-9.
[17]
HIGHAMN J. Functions of Matrices: theory and Computation[M]. Philadelphia: Society for Industrial and Applied Mathematics, 2008.
[18]
LIP, XIEJ, WANGQ, et al. Towards faster training of global covariance pooling networks by iterative matrix square root normalization[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018: 947-955.
[19]
SANNASI CHAKRAVARTHYS R, BHARANIDHARANN, RAJAGURUH. Deep learning-based metaheuristic weighted K-nearest neighbor algorithm for the severity classification of breast cancer[J]. IRBM, 2023, 44(3): 100749.
[20]
KRIZHEVSKYA, SUTSKEVERI, HINTONG E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6): 84-90.
[21]
WANGZ, JIS. Second-order pooling for graph neural networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 45(6): 6870-6880.
[22]
LEXEY DA. An image is worth 16x16 words: transformers for image recognition at scale[EB/OL]. Computer Vision and Pattern Recognition, 2021.
[23]
HEK, ZHANGX, RENS, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016: 770-778.
[24]
HOWARDA, SANDLERM, CHENB, et al. Searching for MobileNetV3[C]//2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019: 1314-1324.
[25]
SIMONYANK, ZISSERMANA. Very deep convolutional networks for large-scale image recognition[EB/OL]. Computer Vision and Pattern Recognition, 2015.arXiv: 1409.1556v6.
[26]
张晋凯. 基于深度学习的乳腺癌组织病理学图像分类研究[D]. 太原: 中北大学, 2023.
[27]
DJOUIMAH, ZITOUNIA, MEGHERBIA C, et al. Classification of breast cancer histopathological images using DensNet201[C]//2022 7th International Conference on Image and Signal Processing and their Applications (ISPA), 2022: 1-6.
[28]
臧敏. 基于深度学习的乳腺癌组织病理图像分类研究[D]. 青岛: 青岛科技大学, 2023.
[29]
CHATTOPADHYAYS, DEYA, SINGHP K, et al. DRDA-Net: Dense residual dual-shuffle attention network for breast cancer classification using histopathological images[J]. Computers in Biology and Medicine, 2022, 145: 105437.