Purpose Breast cancer is the most commonly occurring cancer worldwide. The ultrasound reflectivity imaging technique can be used to obtain breast ultrasound (BUS) images, which can be used to classify benign and malignant tumors. However, the classification is subjective and dependent on the experience and skill of operators and doctors. The automatic classification method can assist doctors and improve the objectivity, but current convolution neural network (CNN) is not good at learning global features and vision transformer (ViT) is not good at extraction local features. In this study, we proposed a visual geometry group attention ViT (VGGA‐ViT) network to overcome their disadvantages. Methods In the proposed method, we used a CNN module to extract the local features and employed a ViT module to learn the global relationship among different regions and enhance the relevant local features. The CNN module was named the VGGA module. It was composed of a VGG backbone, a feature extraction fully connected layer, and a squeeze‐and‐excitation block. Both the VGG backbone and the ViT module were pretrained on the ImageNet dataset and retrained using BUS samples in this study. Two BUS datasets were employed for validation. Results Cross‐validation was conducted on two BUS datasets. For the Dataset A, the proposed VGGA‐ViT network achieved high accuracy (88.71±$\ \pm \ $1.55%), recall (90.73±$\ \pm \ $1.57%), specificity (85.58±$\ \pm \ $3.35%), precision (90.77±$\ \pm \ $1.98%), F1 score (90.73±$\ \pm \ $1.24%), and Matthews correlation coefficient (MCC) (76.34±7$\ \pm \ 7$3.29%), which were better than those of all compared previous networks in this study. The Dataset B was used as a separate test set, the test results showed that the VGGA‐ViT had highest accuracy (81.72±$\ \pm \ $2.99%), recall (64.45±$\ \pm \ $2.96%), specificity (90.28±$\ \pm \ $3.51%), precision (77.08±$\ \pm \ $7.21%), F1 score (70.11±$\ \pm \ $4.25%), and MCC (57.64±$\ \pm \ $6.88%). Conclusions In this study, we proposed the VGGA‐ViT for the BUS classification, which was good at learning both local and global features. The proposed network achieved higher accuracy than the compared previous methods.

Background: The emergence of castration resistance is fatal for patients with prostate cancer (PCa); however, there is still a lack of effective means to detect the early progression. In this study, a novel combined nomogram was established to predict the risk of progression related to castration resistance.Methods: The castration-resistant prostate cancer (CRPC)-related differentially expressed genes (DEGs) were identified by R packages “limma” and “WGCNA” in GSE35988-GPL6480 and GSE70768-GPL10558, respectively. Relationships between DEGs and progression-free interval (PFI) were analyzed using the Kaplan–Meier method in TCGA PCa patients. A multigene signature was built by lasso-penalized Cox regression analysis, and assessed by the receiver operator characteristic (ROC) curve and Kaplan–Meier curve. Finally, the univariate and multivariate Cox regression analyses were used to establish a combined nomogram. The prognostic value of the nomogram was validated by concordance index (C-index), calibration plots, ROC curve, and decision curve analysis (DCA).Results: 15 CRPC-related DEGs were identified finally, of which 13 genes were significantly associated with PFI and used as the candidate genes for modeling. A two-gene (KIFC2 and BCAS1) signature was built to predict the risk of progression. The ROC curve indicated that 5-year area under curve (AUC) in the training, testing, and whole TCGA dataset was 0.722, 0.739, and 0.731, respectively. Patients with high-risk scores were significantly associated with poorer PFI (p < 0.0001). A novel combined nomogram was successfully established for individualized prediction integrating with T stage, Gleason score, and risk score. While the 1-year, 3-year, and 5-year AUC were 0.76, 0.761, and 0.762, respectively, the good prognostic value of the nomogram was also validated by the C-index (0.734), calibration plots, and DCA.Conclusion: The combined nomogram can be used to predict the individualized risk of progression related to castration resistance for PCa patients and has been preliminarily verified to have good predictive ability.