Although using convolutional neural networks (CNNs) for computer-aided diagnosis (CAD) has made tremendous progress in the last few years, the small medical datasets remain to be the major bottleneck in this area. To address this problem, researchers start looking for information out of the medical datasets. Previous efforts mainly leverage information from natural images via transfer learning. More recent research work focuses on integrating knowledge from medical practitioners, either letting networks resemble how practitioners are trained, how they view images, or using extra annotations. In this paper, we propose a scheme named Domain Guided-CNN (DG-CNN) to incorporate the margin information, a feature described in the consensus for radiologists to diagnose cancer in breast ultrasound (BUS) images. In DG-CNN, attention maps that highlight margin areas of tumors are first generated, and then incorporated via different approaches into the networks. We have tested the performance of DG-CNN on our own dataset (including 1485 ultrasound images) and on a public dataset. The results show that DG-CNN can be applied to different network structures like VGG and ResNet to improve their performance. For example, experimental results on our dataset show that with a certain integrating mode, the improvement of using DG-CNN over a baseline network structure ResNet18 is 2.17% in accuracy, 1.69% in sensitivity, 2.64% in specificity and 2.57% in AUC (Area Under Curve). To the best of our knowledge, this is the first time that the margin information is utilized to improve the performance of deep neural networks in diagnosing breast cancer in BUS images.

Due to dramatically increasing information published in social networks, privacy issues have given rise to public concerns. Although the presence of differential privacy provides privacy protection with theoretical foundations, the trade-off between privacy and data utility still demands further improvement. However, most existing studies do not consider the quantitative impact of the adversary when measuring data utility. In this paper, we firstly propose a personalized differential privacy method based on social distance. Then, we analyze the maximum data utility when users and adversaries are blind to the strategy sets of each other. We formalize all the payoff functions in the differential privacy sense, which is followed by the establishment of a static Bayesian game. The trade-off is calculated by deriving the Bayesian Nash equilibrium with a modified reinforcement learning algorithm. The proposed method achieves fast convergence by reducing the cardinality from n to 2. In addition, the in-place trade-off can maximize the user’s data utility if the action sets of the user and the adversary are public while the strategy sets are unrevealed. Our extensive experiments on the real-world dataset prove the proposed model is effective and feasible.

As one of the key operations in Wireless Sensor Networks (WSNs), the energy-efficient data collection schemes have been actively explored in the literature. However, the transform basis for sparsifing the sensed data is usually chosen empirically, and the transformed results are not always the sparsest. In this paper, we propose a Data Collection scheme based on Denoising Autoencoder (DCDA) to solve the above problem. In the data training phase, a Denoising AutoEncoder (DAE) is trained to compute the data measurement matrix and the data reconstruction matrix using the historical sensed data. Then, in the data collection phase, the sensed data of whole network are collected along a data collection tree. The data measurement matrix is utilized to compress the sensed data in each sensor node, and the data reconstruction matrix is utilized to reconstruct the original data in the sink. Finally, the data communication performance and data reconstruction performance of the proposed scheme are evaluated and compared with those of existing schemes using real-world sensed data. The experimental results show that compared to its counterparts, the proposed scheme results in a higher data compression rate, lower energy consumption, more accurate data reconstruction, and faster data reconstruction speed.