Abstract: A processor-implemented method for a neural architecture search (NAS) starts by generating an over-parameterized super network having multiple layers. The super network has multiple operator types. Each of the layers includes a largest super kernel corresponding to a search space. The method also includes performing gradient descent to evolve a largest super kernel to a small kernel corresponding to the search space in order to generate a range of kernel encodings. The method further includes identifying a subset of kernel encodings from the range of kernel encodings, for each layer of the super network, based on the gradient descent. The method determines a set of candidate architectures based on the subset of kernel encodings, each of the candidate architectures having a different model size. The method selects a target model, from the set of architectures, based on meeting hardware specifications, and then applies the target model.

Abstract: An electronic device, a method of determining a memory access efficiency for a memory, and a storage medium are provided, which relate to a field of computer technology, and in particular to fields of chip, memory and processor technologies. The electronic device includes: a memory configured to store executable instructions and data to be processed; and a processor configured to execute the executable instructions so as to at least: read a data block to be tested from the memory; determine a memory access description information according to a size information of the data block to be tested; and determine, according to the memory access description information and a channel description information, a memory access efficiency of the processor in reading the data block to be tested, where the channel description information describes a plurality of channels for the processor to read the data to be processed from the memory.

Abstract: Disclosed herein are approaches for forming dynamic DRAM devices without trench fill voids. A method may include providing a plurality of trenches in a substrate, the plurality of trenches defining a plurality of device structures, and depositing a plurality of layers over the device structures. The layers may include a first layer over the device structures, a second layer over the first layer, and a third layer over the second layer. The method may further include forming a plurality of contact trenches through the plurality of layers to expose one or more device structures of the plurality of device structures, and directing ions into a sidewall of the trenches at a non-zero angle, wherein the ions impact the third layer without impacting the second layer. The method may further include forming a fill material within the trenches after the ions are directed into the sidewall of the trenches.

Abstract: Disclosed is a high-precision dynamic real-time 360-degree omnidirectional point cloud acquisition method based on fringe projection. The method comprises: firstly, by means of the fringe projection technology based on a stereoscopic phase unwrapping method, and with the assistance of an adaptive dynamic depth constraint mechanism, acquiring high-precision three-dimensional (3D) data of an object in real time without any additional auxiliary fringe pattern; and then, after a two-dimensional (2D) matching points optimized by the means of corresponding 3D information is rapidly acquired, by means of a two-thread parallel mechanism, carrying out coarse registration based on Simultaneous Localization and Mapping (SLAM) technology and fine registration based on Iterative Closest Point (ICP) technology.

Abstract: A semiconductor package includes a first semiconductor device and a second semiconductor device. The first semiconductor device includes a first semiconductor substrate, a first bonding structure and a memory cell. The second semiconductor device is stacked over the first semiconductor device. The second semiconductor device includes a second semiconductor substrate, a second bonding structure in a second dielectric layer and a peripheral circuit between the second semiconductor substrate and the second bonding structure. The first bonding structure and the second bonding structure are bonded and disposed between the memory cell and the peripheral circuit, and the memory cell and the peripheral circuit are electrically connected through the first bonding structure and the second bonding structure.

Abstract: The present application provides a method for acquiring aorta based on deep learning and a storage medium, comprising: acquiring a database of slices of an aortic layer and a database of slices of a non-aortic layer; performing deep learning on the database of slices of the aortic layer and the database of slices of the non-aortic layer, to obtain a deep learning model; acquiring CT sequence images to be processed or three-dimensional data of CT sequence images to be processed; extracting feature data from the CT sequence images to be processed or the three-dimensional data of the CT sequence images to be processed; acquiring an image of the aorta from the CT sequence images based on the deep learning model and the feature data.

Abstract: Provided are method and device for testing product, computer device and readable storage medium, wherein in the first testing stage of testing cycle, for each processing machine, the total product number and the tested-product number of products processed by the processing machine are counted; the product testing rate of each processing machine is calculated; in the second testing stage, when machine with product testing rate lower than the preset threshold and machine with the product testing rate higher than the preset threshold appear, products in unprocessed products that meet the preset sampling rule are transferred to the processing machine whose product testing rate is lower than the preset threshold for being processed, and products that do not meet the preset sampling rule are transferred to the processing machine with the product testing rate higher than the preset threshold for being processed, until end of the testing cycle.

Abstract: The present application provides a system for acquiring image of aorta based on deep learning, comprising: a database device, a deep learning device, a data extraction device and an aorta acquisition device; the database device is configured for generating a database of slices of an aorta layer and a database of slices of a non-aorta layer; the deep learning device is connected to the database device, and is configured for performing deep learning on slice data, and for analyzing feature data to obtain aorta data; the data extraction device is configured for extracting feature data of CT sequence images to be processed; the aorta acquisition device is connected to the data extraction device and the deep learning device, and is configured for acquiring an image of aorta from the CT sequence images based on the deep learning model and feature data.

Abstract: The present application provides a method and system for acquiring centerline of aorta based on CT sequence images. The method comprises: acquiring three-dimensional data of CT sequence images; acquiring a gravity center of heart and a gravity center of spine based on the three-dimensional data; filtering impurity data from the three-dimensional data of CT sequence images to obtain an image containing left atrium, left ventricle and without interfering coronary artery tree; layered slicing to obtain a group of binarized images; obtaining a circle center and an radius from each layer of slice in the group of binarized images, to generate a point list and an radius list; and mapping one or more pixel points in the point list and the radius list to the image to obtain a centerline of aorta.

Abstract: An optical switch device and a related method for defining a topological light transport channel in a photonic lattice are provided. An exemplary optical switch device includes a photonic lattice including a photonic topological microring array comprising a plurality of site rings coupled via a plurality of anti-resonant link rings, a plurality of input light ports and a plurality of output light ports located at the plurality of site rings, wherein the plurality of input light ports and the plurality of output light ports are respectively connected by a plurality of topological light transport channels. The optical switch device is further configured such that each of the topological light transport channels is defined by a gain domain area that is produced by a corresponding patterned optical pumping beam emitted onto the photonic topological microring array.

Abstract: Provided is a liquid metal-based flexible electronic device and a method for preparing a liquid metal-based flexible electronic device, that includes: preparing an Acrylonitrile Butadiene Styrene (ABS) plastic model; performing an ion sputtering on a surface of the ABS plastic model to form a gold film, to obtain a gold-plated ABS circuit; introducing a first silica gel into a mold to suspend the gold-plated ABS circuit inside the mold, and curing the first silica gel to obtain a cured model; immersing the cured model in acetone to dissolve the ABS model, to obtain a microchannel with a gold plating on an inner wall of the microchannel in a first silica gel substrate; and injecting a gallium-indium eutectic, inserting a copper wire, and applying a second silica gel and curing the second silica gel, to obtain the liquid metal-based flexible electronic device.

Abstract: An object tracking device includes an object detection unit and a matching unit. The object detection unit includes a plurality of detectors having different processing speeds or accuracy. The object detection unit detects, for each of frames of image data input in time series, each object from the image data using one detector selected from the plurality of the detectors. The matching unit matches a current object detection result which the object detection unit generates from image data of a current frame with a previous object detection result which the object detection unit generates from image data of a previous frame before the current frame, and generates a tracking result of the object based on a matching result.

Abstract: A system is provided for advanced health monitoring and diagnosis based on wearable nano-biosensing networks. Nanophotonic and wireless communication technologies are synergistically leveraged to bridge the gap between nano-biosensing technologies and commercial wearable devices. Embodiments of the presently-disclosed system may include: (1) a nanoplasmonic biochip, implanted subcutaneously and built on a flexible substrate; (2) a nanophotonic smart band or wearable device that is able to collect in-vivo signals on-demand and relay them wirelessly to the user's smartphone by means of a secure data transfer; and (3) advanced signal processing techniques implemented on a remote processor to extract relevant data from the received signals and provide a diagnosis in real-time.

Abstract: A simplified method, an apparatus and a system for measuring coronary artery vascular evaluation parameters are provided.

Abstract: A liquid metal-based flexible electron device and a preparation method are disclosed. In the method, 3D printing and the characteristic that ABS plastic can be dissolved by acetone are utilized, and a microchannel is quickly constructed in the flexible substrate of Ecoflex, and liquid metal is then injected into the microchannel to complete the manufacturing of a flexible electronic device. The gold film on the surface of ABS is transferred to the surface of the flexible Ecoflex substrate.

Abstract: Techniques for identification of global cancer-specific sequence features in whole genome sequence data obtained from cell-free DNA (cfDNA) samples. An exemplary technique includes obtaining a plurality of whole genome sequencing reads from a cfDNA sample and determining two or more metrics from at least a majority of the plurality of genome sequencing reads, where a first metric of the two or more metrics is: (i) a fragment size of the cell free DNA, (ii) relative read depth of the plurality of whole genome sequencing reads, or (iii) germline allelic imbalance. The technique further includes inputting the two or more metrics into a classifier to obtain a first prediction for a first class and a second prediction for a second class, and classifying the sample of cell free DNA as the first class or the second class based on the first prediction and the second prediction.

Abstract: A system is provided for advanced health monitoring and diagnosis based on wearable nano-biosensing networks. Nanophotonic and wireless communication technologies are synergistically leveraged to bridge the gap between nano-biosensing technologies and commercial wearable devices. Embodiments of the presently-disclosed system may include: (1) a nanoplasmonic biochip, implanted subcutaneously and built on a flexible substrate; (2) a nanophotonic smart band or wearable device that is able to collect in-vivo signals on-demand and relay them wirelessly to the user's smartphone by means of a secure data transfer; and (3) advanced signal processing techniques implemented on a remote processor to extract relevant data from the received signals and provide a diagnosis in real-time.

Abstract: A colorimetric array includes a substrate, a first spot on the substrate, and a second spot on the substrate. The first spot includes a first nanoporous pigment that includes a first nanoporous material and a first immobilized, chemoresponsive colorant. The second spot includes a second nanoporous pigment that includes a second nanoporous material and a second immobilized, chemoresponsive colorant. The first nanoporous pigment is different from the second nanoporous pigment.

Abstract: A low stress moisture resistant structure of semiconductor device comprises a low stress moisture resistant layer, wherein a semiconductor device is formed on a semiconductor wafer, the semiconductor device comprises at least one pad, the low stress moisture resistant layer is coated on the semiconductor device and the semiconductor wafer so that a pad top center surface of the pad is exposed. The low stress moisture resistant layer comprises a material comprising crosslinked fluoropolymer. A before-coated stress measured on the semiconductor wafer before the low stress moisture resistant layer is coated and an after-cured stress measured on the semiconductor wafer after the low stress moisture resistant layer is coated and cured define a stress difference, the stress difference is greater than or equal to ?5×107 dyne/cm2 and less than or equal to 5×107 dyne/cm2.

Abstract: The application provides a user interest data analysis and collection system which comprises: a session identification module, which is used for offline extracting log data and a merchandise purchase history record corresponding to the log data from a Web server, and for processing the log data, thereby generating a page browsing sequence reference set; a recommendation service controller which is used for collecting user's browsing record in real time; a log data pre-processing module which is used for, when predetermined conditions are satisfied, receiving the user's browsing log data in this session from the recommendation service controller, and for pre-processing the log data; and a match analysis module which is used for matching the pre-processed log data with the generated page browsing sequence reference set, and determining a set of merchandise that the user is potentially interested in, according to the matching result.