Partial discharge (PD) detection and recognition are of great significance to the condition monitoring of gas‐insulated switchgear (GIS). In the current work, ultra‐high‐frequency (UHF) signals induced by PD current pulses are measured and used to represent PD source. For PD classification, feature parameters need to be extracted from UHF signals. Therefore, this study proposes a new feature extraction method that is based on S‐transform (ST) and singular value decomposition (SVD). PD UHF signals generated by four kinds of artificial defects are collected and analysed. ST is used to acquire the joint time–frequency information of the PD UHF signal. SVD is used to acquire the time–frequency characteristics of the UHF signal. Based on the distribution difference of time–frequency characteristics of different kinds of PDs, a 24‐demensional feature vector is finally extracted. Support vector machine optimised by particle swarm optimisation algorithm is employed as classifier to recognise the four kinds of PDs. Results show that the proposed feature extraction method can effectively identify the designed four kinds of PDs even with few samples and strong background noise.

Background: Clinically, doctors obtain the left ventricular posterior wall thickness (LVPWT) mainly by observing ultrasonic echocardiographic video stream to capture a single frame of images with diagnostic significance, and then mark two key points on both sides of the posterior wall of the left ventricle with their own experience for computer measurement. In the actual measurement, the doctor’s selection point is subjective, and difficult to accurately locate the edge, which will bring errors to the measurement results. Methods: In this paper, a convolutional neural network model of left ventricular posterior wall positioning was built under the TensorFlow framework, and the target region images were obtained after the positioning results were processed by non-local mean filtering and opening operation. Then the edge detection algorithm based on threshold segmentation is used. After the contour was extracted by adjusting the segmentation threshold through prior analysis and the OTSU algorithm, the design algorithm completed the computer selection point measurement of the thickness of the posterior wall of the left ventricle. Results: The proposed method can effectively extract the left ventricular posterior wall contour and measure its thickness. The experimental results show that the relative error between the measurement result and the hospital measurement value is less than 15%, which is less than 20% of the acceptable repeatability error in clinical practice. Conclusions: Therefore, the measurement method proposed in this paper has the advantages of less manual intervention, and the processing method is reasonable and has practical value.

Radiography is an essential basis for the diagnosis of fractures. For the pediatric elbow joint diagnosis, the doctor needs to diagnose abnormalities based on the location and shape of each bone, which is a great challenge for AI algorithms when interpreting radiographs. Bone instance segmentation is an effective upstream task for automatic radiograph interpretation. Pediatric elbow bone instance segmentation is a process by which each bone is extracted separately from radiography. However, the arbitrary directions and the overlapping of bones pose issues for bone instance segmentation. In this paper, we design a detection-segmentation pipeline to tackle these problems by using rotational bounding boxes to detect bones and proposing a robust segmentation method. The proposed pipeline mainly contains three parts: (i) We use Faster R-CNN-style architecture to detect and locate bones. (ii) We adopt the Oriented Bounding Box (OBB) to improve the localizing accuracy. (iii) We design the Global-Local Fusion Segmentation Network to combine the global and local contexts of the overlapped bones. To verify the effectiveness of our proposal, we conduct experiments on our self-constructed dataset that contains 1274 well-annotated pediatric elbow radiographs. The qualitative and quantitative results indicate that the network significantly improves the performance of bone extraction. Our methodology has good potential for applying deep learning in the radiography’s bone instance segmentation.

With the construction of the smart grid, the distribution network with high penetration of the photovoltaic (PV) generator relies more and more on cyber systems to achieve active control; thus, the uncertainty of PV power and the line-switch state will inevitably affect the distribution network. To avoid the situation, a min–max multi-objective two-level planning model is proposed. Firstly, the uncertainty of PV power is considered, and a multi-time PV power model is established. Followed by the analysis of the line-switch state uncertainty in the distribution network, and according to Claude Shannon’s information theory, the line-switch state uncertainty model is established under multiple scenarios. After the distribution network reconfiguration, the Latin hypercube sampling (LHS) method is used to determine the line-switch state when the uncertainty budget is different. Finally, considering the worstcase by the uncertainty of PV power and line-switch status, the control model is proposed to improve the stability of the distribution network with the minimal maintenance cost. The model feasibility is verified by the test system and the characteristics of PV power uncertainty, the line-switch state uncertainty is analyzed, and the influence of the scheduling strategy is discussed, thus providing practical technical support for the distribution network.

Over the past few decades, video quality assessment (VQA) has become a valuable research field. The perception of in-the-wild video quality without reference is mainly challenged by hybrid distortions with dynamic variations and the movement of the content. In order to address this barrier, we propose a no-reference video quality assessment (NR-VQA) method that adds the enhanced awareness of dynamic information to the perception of static objects. Specifically, we use convolutional networks with different dimensions to extract low-level static-dynamic fusion features for video clips and subsequently implement alignment, followed by a temporal memory module consisting of recurrent neural networks branches and fully connected (FC) branches to construct feature associations in a time series. Meanwhile, in order to simulate human visual habits, we built a parametric adaptive network structure to obtain the final score. We further validated the proposed method on four datasets (CVD2014, KoNViD-1k, LIVE-Qualcomm, and LIVE-VQC) to test the generalization ability. Extensive experiments have demonstrated that the proposed method not only outperforms other NR-VQA methods in terms of overall performance of mixed datasets but also achieves competitive performance in individual datasets compared to the existing state-of-the-art methods.

In recent years, the incidence of cerebrovascular diseases (CVD) is increasing, which seriously endangers human health. The study on hemodynamics of cerebrovascular disease can help us to understand, prevent, and treat the disease. As one of the important parameters of human cerebral hemodynamics and tissue metabolism, OEF (oxygen extraction fraction) is of great value in central nervous system diseases. The use of BOLD (blood oxygen level dependent) effect offers the possibility to study cerebral hemodynamic and metabolic characteristics by MRI (magnetic resonance imaging) measurements. Therefore, this paper reviews the hemodynamic parameters of brain tissue, discusses the principles and methods of quantitative BOLD‐based MRI measurements of OEF, and discusses the advantages and disadvantages of each method.