Ultrasound imaging (US) is one of the most commonly used techniques for detecting breast lesions. However, due to the inherent properties of low contrast, speckle noise, and blurred boundaries in B-mode ultrasound images, the performance of breast lesion segmentation was significantly limited. In contrast, original radiofrequency (RF) data contains more detailed information. To enhance the performance of breast lesion segmentation and compensate for the information missing in ultrasound images, this study proposed a radiofrequency and ultrasound image feature attention and fusion network (RUIFAF Net) to comprehensively fuse the complementary information from both modalities. Specifically, we proposed a Feature Attention Module (FAM) and a Multi-level Fusion (MF) method to obtain multi-scale feature information and integrate contextual semantic information, respectively. Experimental results on OASBUD dataset demonstrate that our proposed RUIFAF_Net achieved higher Dice (0.8264 vs. 0.6468-0.8184) and IoU (0.7161 vs. 0.5243-0.7025) compared with the other six advanced methods.

It is important to efficiently and accurately locate the fault root cause to maintain the control performance, when the industrial control system fails. However, this task is very challenging because the industrial control system is large in scale and complex in connection. This paper proposes a novel neural causality analysis network with directed acyclic graph to locate the root cause for complex industrial systems. This network fits the temporal nonlinearity and intervariable non-linearity to mine the causal graph. The proposed method is data-driven, which acts without process knowledge. Compared with the state-of-the-art, this method can effectively output accurate root cause from nonlinear and highly coupled data. The effectiveness and advantages are demonstrated by industrial cases.

Industrial oscillation recordings are often corrupted by underlying nonstationary trend and noisy artifacts, which can occlude features of interest and complicate subsequent oscillation detection and diagnosis. However, there are considerably fewer techniques available in the literature for systematically removing both trend and noise terms from the oscillation measurements. To cater for the general detrending and denoising needs of oscillatory signals, this article proposes an integrated framework featuring the following steps: 1) The ensemble empirical mode decomposition is first adopted to decompose the industrial single-loop data into several intrinsic mode functions (IMFs). 2) To eliminate the trend term, a surrogate-based nonstationarity testing algorithm is implemented to automatically identify and remove the requisite IMFs. 3) By applying canonical correlation analysis on the remained IMFs, the noise-dependent components can be further isolated, which finally yields the detrended and denoised oscillation data. We conducted performance comparison study through extensive simulations and industrial examples. The results demonstrate that the proposed work is a promising tool for industrial oscillation data preprocessing.

Precise measurement of carotid artery blood flow is of vital importance for studying thrombosis and early carotid atherosclerotic plaque. However, the traditional non-parametric methods are limited by the weak detection ability to low-velocity blood flow, and show problems including the large measurement deviation and long algorithm running time. Motivated by the above status quo, a novel method based on deep complex convolutional neural network (DCCNN) is proposed for carotid blood flow velocimetry. Based on supervised learning, DCCNN feeds the echo signals into complex convolutional layers for the purpose of rejecting clutter signals. Then, the outputs of complex convolutional layers are processed by the complex fully connected layers to estimate the blood flow velocity. The effectiveness of the proposed method is verified by simulation as well as in vivo data of healthy volunteers. Compared with typical velocimetry methods such as the high-pass filter and singular value decomposition, the normalized root mean square error (NRMSE) of the velocimetry result obtained from the proposed method is reduced by 47.20%) and 45.45%, and the goodness-of-fit is improved by 5.64%, 3.36%, respectively. In addition, the running time of DCCNN is reduced by 82.10% and 21.11%, respectively. Such results show that the proposed method is a promising tool for blood flow velocity measurement due to its higher velocity measurement accuracy and good real-time performance.