中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

1052 records (took 0.033 seconds)

    Virtual staining systems and methods for observing one or more unstained cells
    212.
    发明授权
    Virtual staining systems and methods for observing one or more unstained cells 有权

    公开(公告)号：US11948295B2

    公开(公告)日：2024-04-02

    申请号：US17256373

    申请日：2019-06-27

    Applicant: Ramot at Tel-Aviv University Ltd.

    Inventor： Natan Tzvi Shaked ,  Alon Shalev ,  Yoav Nahum Nygate

    IPC: G06T7/00 ,  G06T7/13 ,  G06V20/69

    CPC classification number: G06T7/0012 ,  G06T7/13 ,  G06V20/698 ,  G06T2207/10056 ,  G06T2207/20192 ,  G06T2207/30024

    Abstract: Systems and methods for visualizing an unstained sperm cell are presented, the system comprises a data input utility that receives measured data comprising at least one quantitative phase microscopy image of the unstained sperm cell; a data processing utility comprising an image analyzer module that utilizes characteristic refractive index information of one or more organelles of the sperm cell, to process the at least one quantitative phase microscopy image and generate at least one corresponding gradient image that includes edge enhancement of the one or more organelles, and a virtual staining module that applies one or more predetermined virtual staining functions to the at least one quantitative phase microscopy image and at least one corresponding gradient image, thereby virtually stain at least one of the one or more organelles of the sperm cell and generate virtually stained image data of the sperm cell; and an output utility that utilizes the virtually stained image data of the sperm cell and generates one or more stained images of the unstained sperm cell, each stained image emulating an image of the sperm cell should the sperm cell has been actually stained with one or more actual stains.

    MRI IN AN INHOMOGENEOUS FIELD WITH NO PULSED GRADIENTS
    216.
    发明公开
    MRI IN AN INHOMOGENEOUS FIELD WITH NO PULSED GRADIENTS 审中-公开

    公开(公告)号：US20240012078A1

    公开(公告)日：2024-01-11

    申请号：US18021577

    申请日：2021-08-17

    Applicant: Ramot at Tel-Aviv University Ltd.

    Inventor： Uri NEVO ,  David CORCOS

    IPC: G01R33/56 ,  A61B5/055 ,  A61B5/00 ,  G06T7/00

    CPC classification number: G01R33/5608 ,  A61B5/055 ,  A61B5/0042 ,  G06T7/0012 ,  G06T2207/10088 ,  G06T2207/30016 ,  A61B2576/026

    Abstract: A method of scanning an object in a FOV by acquiring an MRI signal from the object at different projections of a spatially encoding magnetic field, using a plurality of receiving antennas, and reconstructing an MRI image of the object, comprising:



    a) at each projection, acquiring the MRI signal from the receiving antennas;
    b) filtering the received signal, for at least some of the projections, by applying different time windows to different components of the signal in different frequency bands, and/or received by different receiver antennas, resulting in a filtered received signal vector whose components describe the filtered received signal as a function of time, at each projection, for one or more receiver antennas; and
    c) reconstructing an image as a vector whose components describe a weighted or unweighted net magnetization at each voxel in the FOV, that would be expected to produce the filtered received signal vector.

    GENERATING AND MANAGING DEEP TENSOR NEURAL NETWORKS
    220.
    发明公开
    GENERATING AND MANAGING DEEP TENSOR NEURAL NETWORKS 审中-公开

    公开(公告)号：US20230306276A1

    公开(公告)日：2023-09-28

    申请号：US17938131

    申请日：2022-10-05

    Applicant: International Business Machines Corporation ,  Trustees of Tufts College ,  Ramot at Tel-Aviv University Ltd.

    Inventor： Lior Horesh ,  Elizabeth Newman ,  Misha E. Kilmer ,  Haim Avron

    IPC: G06N3/126 ,  G06N3/084 ,  G06N3/048

    CPC classification number: G06N3/126 ,  G06N3/084 ,  G06N3/048

    Abstract: Techniques for generating and managing, including simulating and training, deep tensor neural networks are presented. A deep tensor neural network comprises a graph of nodes connected via weighted edges. A network management component (NMC) extracts features from tensor-formatted input data based on tensor-formatted parameters. NMC evolves tensor-formatted input data based on a defined tensor-tensor layer evolution rule, the network generating output data based on evolution of the tensor-formatted input data. The network is activated by non-linear activation functions, wherein the weighted edges and non-linear activation functions operate, based on tensor-tensor functions, to evolve tensor-formatted input data. NMC trains the network based on tensor-formatted training data, comparing output training data output from the network to simulated output data, based on a defined loss function, to determine an update. NMC updates the network, including weight and bias parameters, based on the update, by application of tensor-tensor operations.

Patent Agency Ranking