Abstract: A lithium ion secondary battery and a preparation method therefor, a battery module, a battery pack, and a device are provided. In some embodiments, the lithium ion secondary battery comprises a positive electrode plate, a negative electrode plate and an electrolyte, wherein the positive electrode plate comprises a positive electrode active material and a positive electrode lithium-supplementing material comprising a lithium-rich metal oxide, and the lithium-rich metal oxide comprises one or more elements of Ni, Co, Fe, Mn, and Cu; the electrolyte comprises an electrolyte lithium salt and a solvent, and has a percentage ? of the total mass of a fluorine element in the anions of the electrolyte lithium salt relative to the total mass of the electrolyte of <14%.

Abstract: An impact resistant wound cell comprises a first electrode plate with a first surface and a second surface. The first surface comprises a first active material area and a first current collector area, the second surface comprises a second active material area and a second current collector area. The first current collector area is located on an inner side of an outermost circle of the wound cell. In the first electrode plate, the second current collector area and the first current collector area are adjacent to and opposite to each other, the first current collector area and the second current collector area are bonded together by a first bonding member. The present disclosure further provides a battery using the wound cell.

Abstract: A computer-implemented method to estimate an execution time for large query batches. The method includes identifying a batch of queries, where the batch of queries where each query is configured to retrieve data from a database. Selecting a first query and generating an access path for the first query including one or more mini plans. Building a learning model for the one or more mini plans. Estimating an estimated execution time for each of the one or more mini plans. Adding each of the estimated execution times for each of the one or more mini plans to determine a first total estimated time for the first query. Displaying the first total estimated time.

Abstract: Described herein are neural network-based systems, methods and instrumentalities associated with estimating the motion of an anatomical structure. The motion estimation may be performed using a feature pyramid and/or a motion pyramid that correspond to multiple image scales. The motion estimation may be performed using neural networks and parameters that are learned via a training process involving a student network and a teacher network pre-pretrained with abilities to apply progressive motion compensation.

Abstract: A method for recovering activity of a battery includes obtaining a battery state-of-health value of the battery during a charging process, applying an activation voltage to the battery in response to the battery state-of-health value being less than or equal to a first preset threshold such that lithium ions at a lithium supplement agent of the battery are released to an electrolyte solution, and stopping application of the activation voltage and continuing a normal charge-discharge cycle in response to the battery state-of-health value increasing to be greater than or equal to a second preset threshold. The activation voltage is higher than a normal charging voltage of the battery.

Abstract: This disclosure provides a memory management method and apparatus, to perform refined management on a process access status in a physical host, a virtual machine, or a container, and a traffic occupation status of each control group (cgroup). The method can includes a first record table corresponding to a target cache page is used to record all cgroups that have read or written target data from or into the target cache page and a quantity of times that the target data is read or written by all the cgroups, to improve accuracy of a traffic statistics result of a cgroup. In this case, throttling is performed on a first cgroup based on an updated first record table, so that throttling on the first cgroup can be more fair and accurate.

Abstract: A method includes obtaining n pairs of translation sentences of a source language and a target language, where each of the n pairs of translation sentences includes a source language sentence and a target language sentence that are translations of each other, extracting a source language segment from each source language sentence in the n pairs of translation sentences using an extraction rule of the source language, extracting a target language segment from each target language sentence in the n pairs of translation sentences, and generating an extraction rule of the target language based on n target language segments extracted from n target language sentences.

Abstract: A method for non-invasive quantification of organ fat using a magnetic resonance approach includes: constructing a detection system; connecting a detection area; detection system startup; acquiring data; analyzing data; and performing horizontal data analysis. An external computer, a radio frequency (RF) subsystem, and a portable magnet module are used to construct a system for non-invasive quantification of organ fat based on low-field nuclear magnetic resonance (LF-NMR), which causes no damage, and achieves accurate and non-invasive quantification of organ fat. Specific pulse sequences are used to excite nuclear spin in a target region to generate LF-NMR, so as to achieve “one-click” detection, which is used for fast screening of related diseases such as non-alcoholic fatty liver disease (NAFLD). The system has accurate quantification, and is easy to operate without constraints of operator qualifications.

Abstract: An apparatus for stent visualization includes a hardware processor that is configured to input one or more stent images from a sequence of X-ray images and corresponding balloon marker location data to a cascaded spatial transform network. The background is separated from the one or more stent images using the cascaded spatial transform network and a transformed stent image with a clear background and a non-stent background image is generated. The stent layer and non-stent layer are generated using a neural network without online optimization. A mapping function f maps the inputs, the sequence images and marker coordinates, into the two single image outputs.

Abstract: A computer-implemented method for access path optimization is provided according to embodiments of the present disclosure. In the method, a plurality of real values of an access path factor may be collected during a specified time period. One of the real values may be generated when a query is executed on a first access path. Then, at least one second access path may be generated for the query based on the plurality of real values of the access path factor. Moreover, an optimal access path for the query may be identified from the first access path and the at least one second access path.

Abstract: Described herein are neural network-based systems, methods and instrumentalities associated with estimating the motion of an anatomical structure. The motion estimation may be performed utilizing pre-learned knowledge of the anatomy of the anatomical structure. The anatomical knowledge may be learned via a variational autoencoder, which may then be used to optimize the parameters of a motion estimation neural network system such that, when performing motion estimation for the anatomical structure, the motion estimation neural network system may produce results that conform with the underlying anatomy of anatomical structure.

Abstract: Techniques, systems, and devices are described for providing a computational frame for estimating high-dimensional stochastic behaviors. In one exemplary aspect, a method for performing numerical estimation includes receiving a set of measurements of a stochastic behavior. The set of correlated measurements follows a non-standard probability distribution and is non-linearly correlated. Also, a non-linear relationship exists between a set of system variables that describes the stochastic behavior and a corresponding set of measurements. The method includes determining, based on the set of measurements, a numerical model of the stochastic behavior. The numerical model comprises a feature space comprising non-correlated features corresponding to the stochastic behavior. The non-correlated features have a dimensionality of M and the set of measurements has a dimensionality of N, M being smaller than N.

Abstract: Described herein are systems, methods, and instrumentalities associated with image segmentation such as tubular structure segmentation. An artificial neural network is trained to segment tubular structures of interest in a medical scan image based on annotated images of a different type of tubular structures that may have a different contrast and/or appearance from the tubular structures of interest. The training may be conducted in multiple stages during which a segmentation model learned from the annotated images during a first stage may be modified to fit the tubular structures of interest in a second stage. In examples, the tubular structures of interest may include coronary arteries, catheters, guide wires, etc., and the annotated images used for training the artificial neural network may include blood vessels such as retina blood vessels.

Abstract: Described herein are systems, methods, and instrumentalities associated with automatically detecting and enhancing multiple objects in medical scan images. The detection and/or enhancement may be accomplished utilizing artificial neural networks such as one or more classification neural networks and/or one or more graph neural networks. The neural networks may be used to detect areas in the medical scan images that may correspond to the objects of interest and cluster the areas belonging to a same object into a respective cluster. These tasks may be accomplished, for example, by representing the areas corresponding to the objects of interest and their interrelationships with a graph and processing the graph through the one or more graph neural networks so that the areas belonging to each object may be properly labeled and clustered. The clusters may then be used to enhance the objects of interests in one or more output scan images.

Abstract: Described herein are systems, methods, and instrumentalities associated with denoising medical images such as fluoroscopic images using deep learning techniques. A first artificial neural network (ANN) is trained to denoise an input medical image in accordance with a provided target noise level. The training of the first ANN is conducted by pairing a noisy input image with target denoised images that include different levels of noise. These target denoised images are generated using a second ANN as intermediate outputs of the second ANN during different training iterations. As such, the first ANN may learn to perform the denoising task in an unsupervised manner without requiring noise-free training images as the ground truth.

Abstract: Described herein are systems, methods, and instrumentalities associated with landmark detection. The detection may be accomplished by determining a graph representation of a plurality of hypothetical landmarks detected in one or more medical images. The graph representation may include nodes that represent the hypothetical landmarks and edges that represent the relationships between paired hypothetical landmarks. The graph representation may be processed using a graph neural network such a message passing graph neural network, by which the landmark detection problem may be converted and solved as a graph node labeling problem.

Abstract: An adhesive tape includes a composite membrane and an adhesive layer, where the composite membrane includes a substrate and a hardening layer provided on a surface of the substrate, and a hardness of the hardening layer is 3H to 9H. The adhesive tape provided in this application has both small thickness and high hardness, which increases volumetric energy density of the electrochemical apparatus and can further effectively avoid tape wrinkling of the electrochemical apparatus.

Abstract: Described herein are systems, methods and instrumentalities associated with automatic assessment of aneurysms. An automatic aneurysm assessment system or apparatus may be configured to obtain, e.g., using a pre-trained artificial neural network, strain values associated one or more locations of a human heart and one or more cardiac phases of the human heart and derive a representation (e.g., a 2D matrix) of the strain values across time and/or space. The system or apparatus may determine, based on the derived representation of the strain values, respective strain patterns associated with the one or more locations of the human heart and further determine whether the one or more locations are aneurysm locations by comparing the automatically determined strain patterns with predetermined normal strain patterns of the heart and determining the presence or risk of aneurysms based on the comparison.

Abstract: Described herein are systems, methods, and instrumentalities associated with generating multi-contrast MRI images associated with an MRI study. The systems, methods, and instrumentalities utilize an artificial neural network (ANN) trained to jointly determine MRI data sampling patterns for the multiple contrasts based on predetermined quality criteria associated with the MRI study and reconstruct MRI images with the multiple contrasts based on under-sampled MRI data acquired using the sampling patterns. The training of the ANN may be conducted with an objective to improve the quality of the whole MRI study rather than individual contrasts. As such, the ANN may learn to allocate resources among the multiple contrasts in a manner that optimizes the performance of the whole MRI study.

Abstract: This application discloses a decentralization processing method, which are applied to a microservice system. A first communication proxy in the microservice system receives a first microservice request sent by a first microservice application, the first communication proxy determines, based on a target control plane rule, a microservice access address corresponding to the first microservice request, and sends the first microservice request to the second communication proxy having the microservice access address, where the second communication proxy is used to forward the first microservice request to the second microservice application. The first communication proxy can send the microservice request to the corresponding second communication proxy, and a control plane does not need to route the microservice request, so that traffic for routing the microservice request by the control plane is effectively reduced, and impact of a burst of microservice requests on the control plane is effectively avoided.