Abstract: A deep neural network for metal artifact reduction is described. A method for computed tomography (CT) metal artifact reduction (MAR) includes receiving a first CT image data; receiving a second CT image data; and generating, by an artificial neural network (ANN), CT output image data configured to include fewer artifacts compared to the first and second CT image data. The ANN includes at least two parallel CT image data streams and a CT output image data stream. A first of the at least two parallel CT image data streams is based, at least in part, on the first CT image data, a second of the at least two parallel CT image data stream is based, at least in part, on the second CT image data. The CT output image data stream is based, at least in part, on respective outputs of the at least two parallel CT image data streams.

Abstract: The present disclosure provides an OLED display panel, a preparation method thereof and an OLED display device. By providing an uneven surface on a side of a first encapsulation layer away from the organic light-emitting function layer in a non-display area, the uneven surface of the first encapsulation layer can block the flow of the second encapsulation layer to a certain extent, so as to reduce the fluidity and the climbing distance of the edge of the second encapsulation layer, and increase the edge stress and the slope angle. Thereby the narrow frame design of the product is achieved, the thickness uniformity of the edge of the second encapsulation layer is improved, the Mura defect in the non-display area is avoided, and the encapsulation result is guaranteed.

Abstract: A display substrate and a display device are provided. The display substrate includes: a plurality of sub-pixels; a first electrode layer including a plurality of anode structures; and a pixel definition layer located on a side of the first electrode layer away from the base substrate and including a plurality of openings. For at least some sub-pixels, an orthographic projection of the opening of each sub-pixel falls into that of the anode structure of the sub-pixel on the base substrate, a figure of the orthographic projection of the opening of each sub-pixel has a different shape from a figure of the orthographic projection of the anode structure of the sub-pixel on the base substrate. The figure of the orthographic projection of the anode structure of each sub-pixel has more symmetry axis than the figure of the orthographic projection of the opening of the sub-pixel on the base substrate.

Abstract: A machine-learning-based monochromatic CT image reconstruction method is described for quantitative CT imaging. The neural network is configured to learn a nonlinear mapping function from a training data set to map a CT image, which is reconstructed from a single spectral current-integrating projection data set, to monochromatic projections at a pre-specified energy level, realizing monochromatic CT imaging and overcoming beam hardening. An apparatus, method and/or system are configured to determine, by a trained artificial neural network (ANN), a monochromatic projection data set based, at least in part, on a measured CT image. The measured CT image may be reconstructed based, at least in part, on measured projection data. The measured projection data may be polychromatic. The apparatus, method and/or system may be further configured to reconstruct a monochromatic CT image based, at least in part, on the monochromatic projection data set.

Abstract: A system for generating a high resolution (HR) computed tomography (CT) image from a low resolution (LR) CT image is described. The system includes a first generative adversarial network (GAN) and a second GAN. The first GAN includes a first generative neural network (G) configured to receive a training LR image dataset and to generate a corresponding estimated HR image dataset, and a first discriminative neural network (DY) configured to compare a training HR image dataset and the estimated HR image dataset. The second GAN includes a second generative neural network (F) configured to receive the training HR image dataset and to generate a corresponding estimated LR image dataset, and a second discriminative neural network (DX) configured to compare the training LR image dataset and the estimated LR image dataset.

Abstract: Techniques have been developed to facilitate (1) the capture and pitch correction of vocal performances on handheld or other portable computing devices and (2) the mixing of such pitch-corrected vocal performances with backing tracks for audible rendering on targets that include such portable computing devices and as well as desktops, workstations, gaming stations, even telephony targets. Implementations of the described techniques employ signal processing techniques and allocations of system functionality that are suitable given the generally limited capabilities of such handheld or portable computing devices and that facilitate efficient encoding and communication of the pitch-corrected vocal performances (or precursors or derivatives thereof) via wireless and/or wired bandwidth-limited networks for rendering on portable computing devices or other targets.

Abstract: One embodiment provides a method of training an artificial neural network (ANN) for denoising. The method includes generating, by a similarity module, a respective set of similar elements for each noisy input element of a number of noisy input elements included in a single noisy input data set. Each noisy input element includes information and noise. The method further includes generating, by a sample pair module, a plurality of training sample pairs. Each training sample pair includes a pair of selected similar elements corresponding to a respective noisy input element. The method further includes training, by a training module, an ANN using the plurality of training sample pairs. Each set of similar elements is generated prior to training the ANN. The plurality of training sample pairs is generated during training the ANN. The training is unsupervised.

Abstract: A memory system includes a first controller, a nonvolatile memory connected to the first controller, and a power supply circuit that applies a voltage to the first controller and the nonvolatile memory. The first controller writes first data to a first address in the nonvolatile memory and instructs the power supply circuit to apply the destruction voltage to the nonvolatile memory. The first controller determines whether the first data can be correctly read from the first address in the nonvolatile memory and checks destruction of the nonvolatile memory based on the determination.

Abstract: A system for few-view computed tomography (CT) image reconstruction is described. The system includes a preprocessing module, a first generator network, and a discriminator network. The preprocessing module is configured to apply a ramp filter to an input sinogram to yield a filtered sinogram. The first generator network is configured to receive the filtered sinogram, to learn a filtered back-projection operation and to provide a first reconstructed image as output. The first reconstructed image corresponds to the input sinogram. The discriminator network is configured to determine whether a received image corresponds to the first reconstructed image or a corresponding ground truth image. The generator network and the discriminator network correspond to a Wasserstein generative adversarial network (WGAN). The WGAN is optimized using an objective function based, at least in part, on a Wasserstein distance and based, at least in part, on a gradient penalty.

Abstract: An offshore floating-type wind power combined semi-submersible platform foundation has at least three stand columns that form an enclosure having a polygonal structure. The adjacent stand columns are connected by an upper support and a lower support. Each stand column is composed of an upper column body and a lower column body that are arranged coaxially. An upper portion of the upper column body is provided with a support block, the upper support rests on the support block. A lower portion of the upper column body is provided with a lower support connecting portion that is connected to the lower support. With the structural form of the offshore floating-type wind power combined semi-submersible platform foundation, the requirements for a manufacturing site and a combination site are greatly reduced, providing the possibility of large-scale offsite construction.

Abstract: The present disclosure provides a cooling structure of heat pipe for superconducting magneto plasma dynamic thruster having a cylindrical structure and includes a cathode, an intermediate connector and an anode. The cathode is arranged inside the intermediate connector, the anode is arranged outside the intermediate connector; the cathode is provided with a cathode cooling mechanism, and the anode is provided with an anode cooling mechanism. The cathode cooling mechanism includes a cathode heat pipe and a cathode heat dissipation fin. The anode heat pipe cooling mechanism includes an anode heat pipe and an anode heat dissipation fin.

Abstract: Disclosed are a GLP-1-mimicking pharmaceutical composition for treating type 2 diabetes and a preparation method thereof, wherein the pharmaceutical composition contains polyethylene glycol loxenatide, a physiologically acceptable buffer with the pH of 3.0-7.0, and pharmaceutically acceptable excipients. The pharmaceutical composition has good drug stability.

Abstract: A display substrate, a method for manufacturing a display substrate, and a display device are provided. The display substrate includes an effective display area, a punch area, and a critical area between the effective display area and the punch area. The critical area of the display substrate includes: a substrate, and at least one first barrier wall on the substrate. The at least one first barrier wall is a convex structure.

Abstract: Training a CNN with pseudo ground truth for CT artifact reduction is described. An estimated ground truth apparatus is configured to generate an estimated ground truth image based, at least in part, on an initial CT image that includes an artifact. Feature addition circuitry is configured to add a respective feature to each of a number, N, copies of the estimated ground truth image to create the number, N, initial training images. A computed tomography (CT) simulation circuitry is configured to generate a plurality of simulated training CT images based, at least in part, on at least some of the N initial training images. An artifact reduction circuitry is configured to generate a plurality of input training CT images based, at least in part, on the simulated training CT images. A CNN training circuitry is configured to train the CNN based, at least in part, on the input training CT images and based, at least in part, on the initial training images.

Abstract: Systems and method for performing X-ray computed tomography (CT) that can improve spectral separation and decrease motion artifacts without increasing radiation dose are provided. The systems and method can be used with either a kVp-switching source or a single-kVp source. When used with a kVp-switching source, an absorption grating and a filter grating can be disposed between the X-ray source and the sample to be imaged. Relative motion of the filter and absorption gratings can by synchronized to the kVp switching frequency of the X-ray source. When used with a single-kVp source, a combination of absorption and filter gratings can be used and can be driven in an oscillation movement that is optimized for a single-kVp X-ray source. With a single-kVp source, the absorption grating can also be omitted and the filter grating can remain stationary.

Abstract: A neural network based corrector for photon counting detectors is described. A method for photon count correction includes receiving, by a trained artificial neural network (ANN), a detected photon count from a photon counting detector. The detected photon count corresponds to an attenuated energy spectrum. The attenuated energy spectrum is related to characteristics of an imaging object and is based, at least in part, on an incident energy spectrum. The method further includes correcting, by the trained ANN, the detected photon count to produce a corrected photon count. The method may include reconstructing, by image reconstruction circuitry, an image based, at least in part, on the corrected photon count.

Abstract: A system for enhancing a low-dose (LD) computed tomography (CT) image is described. The system includes a modularized adaptive processing neural network (MAP-NN) apparatus and a MAP module. The MAP-NN apparatus is configured to receive a LDCT image as input. The MAP-NN apparatus includes a number, T, trained neural network (NN) modules coupled in series. Each trained NN module is configured to generate a respective test intermediate output image based, at least in part, on a respective received test input image. Each test intermediate output image corresponds to an incrementally denoised respective received test input image. The MAP module is configured to identify an optimum mapping depth, D, based, at least in part, on a selected test intermediate output image, the selected test intermediate output image selected by a domain expert. The mapping depth, D, is less than or equal to the number, T.

Abstract: A deep neural network for metal artifact reduction is described. A method for computed tomography (CT) metal artifact reduction (MAR) includes receiving a first CT image data; receiving a second CT image data; and generating, by an artificial neural network (ANN), CT output image data configured to include fewer artifacts compared to the first and second CT image data. The ANN includes at least two parallel CT image data streams and a CT output image data stream. A first of the at least two parallel CT image data streams is based, at least in part, on the first CT image data, a second of the at least two parallel CT image data stream is based, at least in part, on the second CT image data. The CT output image data stream is based, at least in part, on respective outputs of the at least two parallel CT image data streams.

Abstract: In one embodiment, there is provided a dual neck autoencoder module for reducing adversarial attack transferability. The dual neck autoencoder module includes an encoder module configured to receive input data; a decoder module; and a first bottleneck module and a second bottleneck module coupled, in parallel, between the encoder module and the decoder module. The decoder module is configured to generate a first estimate based, at least in part, on a first intermediate data set from the first bottleneck module, and a second estimate based, at least in part, on a second intermediate data set from the second bottleneck module. The first intermediate data set and the second intermediate data set are at least partially decorrelated based, at least in part, on a correlation loss.

Abstract: A display substrate, a method for manufacturing a display substrate, and a display device are provided. The display substrate includes an effective display area, a punch area, and a critical area between the effective display area and the punch area. The critical area of the display substrate includes: a substrate, and at least one first barrier wall on the substrate. The at least one first barrier wall is a convex structure.