Abstract: A charging management method for multiple battery packs comprises determining a target charging current of each battery pack according to a sampled temperature of each battery pack and a sampled voltage of each battery pack; calculating a difference between the sum of the target charging currents of various battery packs and the sum of sampled charging currents of the various battery packs, to obtain a total PI adjustment value; sending the total PI adjustment value to a charging device, to enable the charging device to adjust a total charging current according to the total PI adjustment value; determining a charging current of each battery pack according to the sampled charging current of each battery pack and the total charging current, and determining a charging voltage of each battery pack according to the sampled voltage of each battery pack.

Abstract: A method for quantitative magnetic resonance imaging (MRI) includes [800] performing an MRI scan using a conventional pulse sequence to obtain a qualitative MR image; and [802] applying the qualitative MR image as input to a deep convolutional neural network (CNN) to produce a quantitative magnetic resonance (MR) relaxation parametric map. The qualitative MR image is the only image input to the deep neural network to produce the quantitative MR relaxation parametric map. The conventional pulse sequence may be a Spoiled Gradient Echo sequence, a Fast Spin Echo sequence, a Steady-State Free Precession sequence, or other sequence that is commonly used in current clinical practice.

Abstract: The present invention provides a selection method of array length of observation system, comprising: step S101: obtaining a seismic response equation for any point on ground based on wavefield propagation theory; step S102: determining at least one selection criterion of an optimum array length considering different factors according to the seismic response equation; step S103: obtaining the array lengths considering the different factors respectively according to the at least one selection criterion of the optimum array length; and step S104: integrating the array lengths considering the different factors and determining that the optimum array length is ?{square root over (2)} times the depth of the destination layer. Compared with conventional array length calculation method the demonstration method of array length as proposed in the present invention has the advantages of higher accuracy and applicability for destination layers.

Abstract: The present disclosure provides a universal preparation method for in-situ growth of a layered double hydroxide (LDH) layer on a substrate surface, and belongs to the technical field of material synthesis. In the present disclosure, an LDH protective layer is grown in situ on a surface of a substrate by means of electrodeposition combined with hydrothermal treatment. Specifically, a seed crystal layer of the LDH is formed on the substrate surface by the electrodeposition, and then obtained LDH seed crystals are crystallized and grown by Ostwald ripening through the hydrothermal treatment. In this way, the LDH protective layer is formed in which an interlayer anion is a nitrate. The protective layer protects the substrate against corrosion. Moreover, since the interlayer anion is the nitrate, the protective layer can be exchanged with other corrosion-inhibiting anions, and is modifiable.

Abstract: At least one processor identifies dependency relationships among libraries in a repository of libraries. Using the dependency relationships among libraries, at least one machine learning model can be created that predicts with a confidence value a dependency between a given library and a target library. An L layer tree-like graph can be created, using the dependency relationships among libraries and an application package. L can be configurable. Versions of the libraries to use can be determined by running the at least one machine learning model for each pair of nodes having a dependency relationship in the L layer tree-like graph, the at least one machine learning model identifying the dependency relationship with a confidence value, where pairs of nodes having largest confidence values are selected as the versions of the libraries to use in the application package.

Abstract: A thermoset component for joining with a thermoplastic component according to an exemplary embodiment of this disclosure, among other possible things includes a thermoset material, a weldable surface on the thermoset material, and a thermal barrier. The weldable surface includes a thermoplastic material. A hybrid polymeric component and a method of making a hybrid polymeric component are also disclosed.

Abstract: A method for medical imaging performs a sparse-sampled tomographic imaging acquisition by an imaging system to produce acquired sparse imaging samples; synthesizes by a first deep learning network unacquired imaging samples from the acquired imaging samples to produce complete imaging samples comprising both the acquired imaging samples and unacquired imaging samples; transforms by a physics module the complete imaging samples to image space data based on physics and geometry priors of the imaging system; and performs image refinement by a second deep learning network to produce tomographic images from the image space data. The physics and geometry priors of the imaging system comprise geometric priors of a physical imaging model of the imaging system, and prior geometric relationships between the sample and image data domains.

Abstract: A treatment planning prediction method to predict a Dose-Volume Histogram (DVH) or Dose Distribution (DD) for patient data using a machine-learning computer framework is provided with the key inclusion of a Planning Target Volume (PTV) only treatment plan in the framework. A dosimetric parameter is used as an additional parameter to the framework and which is obtained from a prediction of the PTV-only treatment plan. The method outputs a Dose-Volume Histogram and/or a Dose Distribution for the patient including the prediction of the PTV-only treatment plan. The method alleviates the complicated process of quantifying anatomical features and harnesses directly the inherent correlation between the PTV-only plan and the clinical plan in the dose domain. The method provides a more robust and efficient solution to the important DVHs prediction problem in treatment planning and plan quality assurance.

Abstract: A transformer bobbin includes a first bobbin member including a first end defining a first opening, a second end defining a second opening, an inner surface defining the first opening and an outer surface. The inner surface includes a first plurality of fins that extend between the first end and the second end. A second bobbin member is disposed about and spaced from the first bobbin member. The second bobbin member includes a first end portion defining a first opening portion, a second end portion, an inner surface portion and an outer surface portion. One of the outer surface of the first bobbin member and the outer surface portion of the second bobbin member includes a second plurality of fins that extend between the first end portion and the second end portion.

Abstract: A method for radiation therapy treatment planning includes: a) defining a decision variable search space by projecting an initial seed point onto a pareto front, where the pareto front and decision variable search space are defined in the same coordinate space; b) projecting search points in the decision variable search space to the pareto front using a one-dimensional search algorithm to produce projected points on the pareto front; c) updating the search points by performing a gradient-free optimization that evaluates a treatment planning scoring function at the projected points on the pareto front; and d) repeating steps (b), (c) to search within the decision variable search space until a convergence criterion is satisfied, thereby producing a pareto optimal and clinically acceptable treatment plan.

Abstract: Retrospective magnetic resonance imaging (MRI) uses a deep neural network framework [102] to generate from MRI imaging data [100] acquired by an MRI apparatus using a predetermined imaging protocol tissue relaxation parametric maps and magnetic/radiofrequency field maps [104] which are then used to generate using the Bloch equations [106] predicted MRI images [108] corresponding to imaging protocols distinct from the predetermined imaging protocol. This allows obtaining a wide spectrum of tissue contrasts distinct from those of the acquired MRI imaging data.

Abstract: The disclosure provides a precursor solution, a perovskite solar cell and a preparation method thereof. The solute of the precursor solution includes a metal halide, the solvent of the precursor solution is an organic solvent, and the precursor solution contains nanobubbles, which have a diameter not more than 1000 nm, and the zeta potential of the precursor solution does not exceed ?20 mV. The method of preparing the precursor solution includes: (1) preparing an organic solvent containing nanobubbles; (2) dissolving a solute in the organic solvent containing nanobubbles. The precursor solution of the disclosure has a very low zeta potential, and the nanobubbles can exist stably in the organic solvent(s) for up to one month. When comparing with traditional methods for preparing the precursor solution of the perovskite cells, the method for preparing the precursor solution of the disclosure can effectively improve the stability, reproducibility and solubility of the metal halide in the organic solvent(s).

Abstract: A method for diagnostic imaging reconstruction uses a prior image xpr from a scan of a subject to initialize parameters of a neural network which maps coordinates in image space to corresponding intensity values in the prior image. The parameters are initialized by minimizing an objective function representing a difference between intensity values of the prior image and predicted intensity values output from the neural network. The neural network is then trained using subsampled (sparse) measurements of the subject to learn a neural representation of a reconstructed image. The training includes minimizing an objective function representing a difference between the subsampled measurements and a forward model applied to predicted image intensity values output from the neural network. Image intensity values output from the trained neural network from coordinates in image space input to the trained neural network are computed to produce predicted image intensity values.

Abstract: Image reconstruction is an inverse problem that solves for a computational image based on sampled sensor measurement. Sparsely sampled image reconstruction poses addition challenges due to limited measurements. In this work, we propose an implicit Neural Representation learning methodology with Prior embedding (NeRP) to reconstruct a computational image from sparsely sampled measurements. The method differs fundamentally from previous deep learning-based image reconstruction approaches in that NeRP exploits the internal information in an image prior, and the physics of the sparsely sampled measurements to produce a representation of the unknown subject. No large-scale data is required to train the NeRP except for a prior image and sparsely sampled measurements. In addition, we demonstrate that NeRP is a general methodology that generalizes to different imaging modalities such as CT and MRI.

Abstract: Retrospective magnetic resonance imaging (MRI) uses a deep neural network framework [102] to generate from MRI imaging data [100] acquired by an MRI apparatus using a predetermined imaging protocol tissue relaxation parametric maps and magnetic/radiofrequency field maps [104] which are then used to generate using the Bloch equations [106] predicted MRI images [108] corresponding to imaging protocols distinct from the predetermined imaging protocol. This allows obtaining a wide spectrum of tissue contrasts distinct from those of the acquired MRI imaging data.

Abstract: A joint between dissimilar thermoplastic materials comprising a first thermoplastic material layer; a second thermoplastic material layer having a melting point temperature different from a melting point temperature of the first thermoplastic material layer; and an interface layer coupled between the first thermoplastic material layer and the second thermoplastic material layer; wherein the interface layer is configured to join the first thermoplastic material layer and the second thermoplastic material layer together to form the joint, wherein the interface layer comprises a melting point temperature having a value selected from the group consisting of between the melting point temperature of the first thermoplastic material layer and the melting point temperature of the second thermoplastic material layer; or lower than the melting point temperature of the first thermoplastic material layer and the melting point temperature of the second thermoplastic material layer.

Abstract: Over 2 million cystoscopies are performed annually in the United States and Europe for detection and surveillance of bladder cancer. Adequate identification of suspicious lesions is critical to minimizing recurrence and progression rates, however standard cystoscopy misses up to 20% of bladder cancer. Access to adjunct imaging technology may be limited by cost and availability of experienced personnel. Machine learning holds the potential to enhance medical decision-making in cancer detection and imaging. Various embodiments described herein are directed to methods for identifying cancers, tumors, and/or other abnormalities present in a person's bladder. Additional embodiments are directed to machine learning systems to identify cancers, tumors, and/or other abnormalities present in a person's bladder, while additional embodiments will also identify benign or native structures or features in a person's bladder.

Abstract: Provided is a salt of an Syk inhibitor and a crystalline form thereof, and more specifically, provided are 5-fluoro-1-methyl-3-((5-(4-(oxetan-3-yl)piperazine-1)-yl)pyridin-2-yl)amino)-6-(1H-pyrazol-3-yl)quinoline-2(1H)-ketamine hydrochloride, a crystalline form thereof, a preparation method therefor, a pharmaceutical composition thereof and a use thereof. The hydrochloride of the compound represented by formula I and the crystalline form thereof have good salt-forming properties, high stability and low hygroscopicity, have advantages in terms of physical properties, safety and metabolic stability, and have value in prepared medicines.

Abstract: A method for radiation therapy treatment planning includes: a) defining a decision variable search space by projecting an initial seed point onto a pareto front, where the pareto front and decision variable search space are defined in the same coordinate space; b) projecting search points in the decision variable search space to the pareto front using a one-dimensional search algorithm to produce projected points on the pareto front; c) updating the search points by performing a gradient-free optimization that evaluates a treatment planning scoring function at the projected points on the pareto front; and d) repeating steps (b), (c) to search within the decision variable search space until a convergence criterion is satisfied, thereby producing a pareto optimal and clinically acceptable treatment plan.

Abstract: A method for tomographic imaging comprising acquiring [200] a set of one or more 2D projection images [202] and reconstructing [204] a 3D volumetric image [216] from the set of one or more 2D projection images [202] using a residual deep learning network comprising an encoder network, a transform module and a decoder network, wherein the reconstructing comprises: transforming [206] by the encoder network the set of one or more 2D projection images [202] to 2D features [208]; mapping [210] by the transform module the 2D features [208] to 3D features [212]; and generating [214] by the decoder network the 3D volumetric image from the 3D features [212]. Preferably, the encoder network comprises 2D convolution residual blocks and the decoder network comprises 3D blocks without residual shortcuts within each of the 3D blocks.