Breast cancer is one of most serious malignant tumors that affect women’s health. To carry out the early screening of breast cancer, mammography provides breast cancer images for doctors’ efficient diagnosis. However, breast cancer lumps can vary in size and shape, bringing difficulties for the accurate recognition of both humans and machines. Moreover, the annotation of such images requires expert medical knowledge, which increases the cost of collecting datasets to boost the performance of deep learning methods. To alleviate these problems, we propose an attention-based active learning framework for breast cancer segmentation in mammograms; the framework consists of a basic breast cancer segmentation model, an attention-based sampling scheme and an active learning strategy for labelling. The basic segmentation model performs multi-scale feature fusion and enhancement on the basis of UNet, thus improving the distinguishing representation capability of the extracted features for further segmentation. Afterwards, the proposed attention-based sampling scheme assigns different weights for unlabeled breast cancer images by evaluating their uncertainty with the basic segmentation model. Finally, the active learning strategy selects unlabeled images with the highest weights for manual labeling, thus boosting the performance of the basic segmentation model via retraining with new labeled samples. Testing on four datasets, experimental results show that the proposed framework could greatly improve segmentation accuracy by about 15% compared with an existing method, while largely decreasing the cost of data annotation.

Computational ghost imaging is a novel technique, which has a wide range of applications in many fields. As a key part of computational ghost imaging, the measurement matrix plays an important role in imaging quality and system practicability. To improve the imaging quality of computational ghost imaging and overcome the poor stability and non-negativity of the measurement matrix, we propose a new construction method for the deterministic measurement matrix based on the difference set modulo subgroup. This method uses an efficient simulated annealing algorithm to search for the difference set modulo subgroup. Then a 0-1 binary measurement matrix can be constructed according to the obtained difference set. We can show that the measurement matrix constructed by this method has low coherence and can satisfy the restricted isometry property. The simulation and experimental results showed that the reconstruction quality of the proposed measurement matrix was equivalent to the Sparse random matrix and better than the Toeplitz and Circulant matrices, which indicates the feasibility of the newly proposed measurement matrix.

The U-Former model is proposed in this work to segment the COVID-19 lung computed tomography images of patients. U-Former introduces the transformer architecture, based on the traditional U-Net segmentation network, which effectively improves the network’s ability to capture global features. The mixed module is presented in this work to capture long-range dependencies and extract local information. In the mixed module, the computationally expensive self-attention mechanism is enhanced and combined with convolution to enable the network to efficiently capture global information while taking into account local details. The multi-scale attention module is utilized to fuse the multi-scale features to enhance the segmentation effect for details. Experimental results show that the proposed U-Former model outperforms other state-of-the-art segmentation models, including both convolutional neural network-based and transformer-based models, with a mean Dice score of 82.54%, a mean intersection over union of 80.01%, and a mean sensitivity of 85.70%. The code and models are publicly available at https://github.com/tianyuzhou668/U-Former

Spectral clustering is a widely used unsupervised clustering algorithm that performs very well in many cases. However, for complex scenes and high-resolution images, the application is limited due to the high computational complexity. In this paper, we propose an efficient spectral clustering algorithm based on wavelet basis decomposition. According to the hierarchical structure of wavelet decomposition, the algorithm reduces the dimension of the eigendecomposition of the graph Laplacian by wavelet basis matrix, the low-frequency eigenvectors of the whole graph Laplacian are solved hierarchically from wavelet subspaces with different levels. Its computational complexity is $O(n) + O(m3/2)$ , where $n$ and m. are the number of pixels and selected wavelet coefficients in an image, respectively. To verify the effectiveness and performance of the proposed algorithm, a series of experiments were done on both the Weizmann and BSDS500 segmentation datasets and find that our method, which in practice provides on average about 5 × speed-up to the eigendecomposition computation required for the Laplacian matrix with comparable segmentation accuracy.