Abstract: A scheme for modifying an image is disclosed, which includes receiving a source image having a first image configuration; determining a second image configuration for a target image; providing the received source image to an AI engine trained to identify, based on a set of rules related to visual features, candidate regions from the source image; generating proposal images based on the candidate regions, respectively; determining, based on prior aesthetical evaluation data, an aesthetical value of each regional proposal image; and selecting, based on the determined aesthetical value of each regional proposal image, one of the regional proposal images as the target image; extracting, from the AI engine, the target image; and causing the target image to be displayed via a display of a user device.

Abstract: Provided are a backlight module, a preparation method thereof, and a display device. The backlight module includes a substrate, multiple light-emitting elements, a reflection structure and a rubber frame structure, and the multiple light-emitting elements, the reflection structure and the rubber frame structure are located on one side of the substrate. The multiple light-emitting elements and the reflection structure are located within a limited area of the rubber frame structure, and the reflection structure is located between every two adjacent light-emitting elements; and the reflection structure includes a first reflection surface adjacent to a light-emitting element, and the rubber frame structure includes a second reflection surface adjacent to the light-emitting element.

Abstract: The present application relates to a current collector, an electrode plate, a battery cell, a battery, and an electric device. The current collector includes a stainless steel current-collecting portion for collecting a current, where the stainless steel current-collecting portion includes a blank area and a coating area that are arranged adjacent to each other in a preset direction, the coating area is configured to be coated with an active material, and the blank area has a hardness less than that of the coating area.

Abstract: Disclosed are a device and method for detecting defects of a high-voltage cable cross-bonded grounding system. The method comprises: selecting a protective grounding box of a cross-bonded grounding system, respectively connecting a signal excitation coupler to A-phase, B-phase and C-phase coaxial cables of the protective grounding box, selecting a stable signal with a frequency different from a power frequency or a field interference frequency, and testing effective current values and phases responded by the A-phase, B-phase and C-phase coaxial cables when the stable signal with the frequency F1 is injected into the A-phase, B-phase and C-phase coaxial cables of the protective grounding box in a coupled manner; and obtaining resistances and inductances of branch circuits of the cable cross-bonded grounding system by calculation according to measurement data, and determining if the cable cross-bonded grounding system has a connection defect.

Abstract: Motion contaminated magnetic resonance (MR) images for training an artificial neural network to remove motion artifacts from the MR images are difficult to obtain. Described herein are systems, methods, and instrumentalities for injecting motion artifacts into clean MR images and using the artificially contaminated images for machine learning and neural network training. The motion contaminated MR images may be created based on clean source MR images that are associated with multiple physiological cycles of a scanned object, and by deriving MR data segments for the multiple physiological cycles based on the source MR images. The MR data segments thus derived may be combined to obtain a simulated MR data set, from which one or more target MR images may be generated to exhibit a motion artifact. The motion artifact may be created by manipulating the source MR images and/or controlling the manner in which the MR data set or the target MR images are generated.

Abstract: Disclosed is an unmanned aerial vehicle (UAV)-aided over-the-air computing system based on full-duplex relay and a trajectory and power optimization method thereof. The UAV-aided over-the-air computing system adopts the UAV as a full-duplex relay for data fusing and forwarding. The method of the application aims at minimizing an average mean square error of over-the-air computing, and solves an optimization problem under constraints of sensor transmitting power, information transmission rate, UAV trajectory and denoising factor. For an overall joint optimization problem, each optimization variable is determined one by one by an alternative optimization method according to the high coupling of optimization variables.

Abstract: An anatomy-aware contouring editing method includes receiving an image, wherein the image represents an anatomically recognizable structure; identifying a first image segment representing part of the anatomically recognizable structure; annotating the first image segment to generate a label of the part; drawing a contour along a boundary of the part; receiving a first input from a user device indicative of a region of contour failure, wherein the region of contour failure includes a portion of a contour that requires editing; editing the contour for generating an edited contour based on the first input and anatomical information; and updating another contour of another part of the anatomically recognizable structure based on the edited contour, wherein the another part is anatomically related to the part.

Abstract: Methods, computer systems, and apparatus, including computer programs encoded on computer storage media, for processing a perspective view range image generated from sensor measurements of an environment. The perspective view range image includes a plurality of pixels arranged in a two-dimensional grid and including, for each pixel, (i) features of one or more sensor measurements at a location in the environment corresponding to the pixel and (ii) geometry information comprising range features characterizing a range of the location in the environment corresponding to the pixel relative to the one or more sensors. The system processes the perspective view range image using a first neural network to generate an output feature representation. The first neural network comprises a first perspective point-set aggregation layer comprising a geometry-dependent kernel.

Abstract: Disclosed herein are systems, methods, and instrumentalities associated with reconstructing magnetic resonance (MR) images based on multi-slice, under-sampled MRI data (e.g., k-space data). The multi-slice MRI data may be acquired using a simultaneous multi-slice (SMS) technique and MRI information associated with multiple MRI slices may be entangled in the multi-slice MRI data. A neural network may be trained and used to disentangle the MRI information and reconstruct MRI images for the different slices. A data consistency component may be used to estimate k-space data based on estimates made by the neural network, from which respective MRI images associated with multiple MRI slices may be obtained by applying a Fourier transform to the k-space data.

Abstract: A 3D anatomical model of one or more blood vessels of a patient may be obtained using CT angiography, while a 2D image of the blood vessels may be obtained based on fluoroscopy. The 3D model may be registered with the 2D image based on a contrast injection site identified on the 3D model and/or in the 2D image. A fused image may then be created to depict the overlaid 3D model and 2D image, for example, on a monitor or through a virtual reality headset. The injection site may be determined automatically or based on a user input that may include a bounding box drawn around the injection site on the 3D model, a selection of an automatically segmented area in the 3D model, etc.

Abstract: Digital breast tomosynthesis (DBT) may provide richer information than full-field digital mammography (FFDM). DBT data such as DBT slices may be processed based on deep learning techniques such as using a neural network, and the DBT slices may be divided into groups and a pre-determined number of representative images may be derived based on the grouping. The neural network may be configured to process the representative images to predict the presence or non-presence of a breast disease such as breast cancer.

Abstract: Mammography data such as DBT and/or FFDM images may be processed using deep learning based techniques, but labeled training data that may facilitate the learning may be difficult to obtained. Described herein are systems, methods, and instrumentalities associated with automatically generating and/or augmenting labeled mammography training data, and training a deep learn model based on the auto-generated/augmented data for detecting a breast disease (e.g., breast cancer) in a mammography image.

Abstract: A method of operating a vibrational energy harvesting system includes providing an energy harvester and an energy storage module. The energy storage module is coupled to the energy harvester to transfer energy to the energy storage module. The method also includes determining whether the energy transferred exceeds a predetermined threshold. The method further includes limiting the energy transferred to the energy storage module by a passive device when the energy transferred exceeds the predetermined threshold.

Abstract: The composition includes a vinyl-substituted polysiloxane having at least two vinyl groups, a hydrosilylation catalyst, a hydrosilyl-substituted polysiloxane having at least two silicon-hydride groups, and a phosphorous-containing flame retardant encapsulated in a crosslinked, nitrogen-containing polymer. The composition may be a two-part composition including a first part and a second part. The first part includes a vinyl-substituted polysiloxane, a hydrosilylation catalyst, and a phosphorous-containing flame retardant encapsulated in a crosslinked, nitrogen-containing polymer. The second part includes a second vinyl-substituted polysiloxane and a hydrosilyl-substituted polysiloxane. A battery module including a plurality of battery cells and the composition at least partially encasing the plurality of battery cells is also disclosed.

Abstract: An adhesive composition contains an acid anhydride group-containing oil or fat (A) and a curing agent (B) having a reactive group reactive with an acid anhydride group. A laminate is obtained by placing an adhesive layer between a first substrate and a second substrate, and the adhesive layer includes the aforementioned adhesive composition. A package is obtained by forming a bag from a laminate obtained by placing an adhesive layer between a first substrate and a second substrate, and the adhesive layer includes the aforementioned adhesive composition.

Abstract: Embodiments of the present disclosure disclose a method and apparatus for determining a target anchor, a device and a storage medium. The method may include: extracting a plurality of feature maps of an original image using a feature extraction network; inputting the plurality of feature maps into a feature pyramid network to perform feature fusion, to obtain a plurality of fused feature maps; and using a region proposal network to implement operations as follows: determining an initial anchor of a network head using the fused feature map, based on a size of each fused feature map, and determining an offset parameter of the initial anchor, based on a ratio of the size of the fused feature map to the original image, and generating a plurality of candidate anchors in different directions, based on the offset parameter of the initial anchor.

Abstract: A planar antenna can include one or more high permeability objects that are added near the planar coil's tips. Further, the planar antenna includes a coil having two or more layers, with each layer having two or more turns. The planar antenna also has a restricted conductor arrangement, where there are a maximum number of turns arranged in any horizontal layer. A turn's conductor could be a single conductor or could be two (2) or more conductors arranged in parallel.

Abstract: An adaptive prefetcher for a shared system cache of a processing system including multiple requestors having a cache miss monitor and a prefetch controller. The cache miss monitor monitors requests for information from memory and identifies one of the requestors for which an identified cache line is requested. The prefetch controller submits an adaptive request for a subsequent cache line. The subsequent cache line may be determined based on a latency comparison between a loop latency (LL) of the prefetch controller and a stream latency (SL) of the identified requestor. A latency memory may be included that stores stream latencies for the requestors. The latency comparison may be used to determine how many cache lines to skip relative to the identified cache line, such as according to SL*SK<LL?SL*(SK+1) in which SK is the number of cache lines to skip.

Abstract: A scheme for modifying an image is disclosed, which includes receiving a source image having a first image configuration; determining a second image configuration for a target image; providing the received source image to an AI engine trained to identify, based on a set of rules related to visual features, candidate regions from the source image; generating proposal images based on the candidate regions, respectively; determining, based on prior aesthetical evaluation data, an aesthetical value of each regional proposal image; and selecting, based on the determined aesthetical value of each regional proposal image, one of the regional proposal images as the target image; extracting, from the AI engine, the target image; and causing the target image to be displayed via a display of a user device.

Abstract: An adhesive composition contains an acid anhydride group-containing oil or fat (A) and a curing agent (B) having a reactive group reactive with an acid anhydride group. A laminate is obtained by placing an adhesive layer between a first substrate and a second substrate, and the adhesive layer includes the aforementioned adhesive composition. A package is obtained by forming a bag from a laminate obtained by placing an adhesive layer between a first substrate and a second substrate, and the adhesive layer includes the aforementioned adhesive composition.