中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

45 records (took 0.02 seconds)

    Annotation pipeline for machine learning algorithm training and optimization
    33.
    发明授权
    Annotation pipeline for machine learning algorithm training and optimization 有权

    公开(公告)号：US11475358B2

    公开(公告)日：2022-10-18

    申请号：US16527965

    申请日：2019-07-31

    Applicant: GE Precision Healthcare LLC

    Inventor： Marc T. Edgar ,  Travis R. Frosch ,  Gopal B. Avinash ,  Garry M. Whitley

    IPC: G06N20/00 ,  G06N5/04 ,  G16H50/20 ,  G06F40/169 ,  G06K9/62

    Abstract: Techniques are provided for enhancing the efficiency and accuracy of annotating data samples for supervised machine learning algorithms using an advanced annotation pipeline. According to an embodiment, a method can comprise collecting, by a system comprising a processor, unannotated data samples for input to a machine learning model and storing the unannotated data samples in an annotation queue. The method further comprises determining, by the system, annotation priority levels for respective unannotated data samples of the unannotated data samples, selecting, by the system from amongst different annotation techniques, one or more of the different annotation techniques for annotating the respective unannotated data samples based the annotation priority levels associated with the respective unannotated data samples.

    DIFFERENTIAL LEARNING FOR LEARNING NETWORKS
    34.
    发明申请
    DIFFERENTIAL LEARNING FOR LEARNING NETWORKS 有权

    公开(公告)号：US20210374513A1

    公开(公告)日：2021-12-02

    申请号：US16883315

    申请日：2020-05-26

    Applicant: GE Precision Healthcare LLC

    Inventor： Chiranjib Sur ,  Venkata Ratnam Saripalli ,  Gopal B. Avinash

    IPC: G06N3/08 ,  G06N3/04 ,  G16H30/40 ,  G16H50/20 ,  G06T7/00

    Abstract: A computer-implemented system is provided that includes a learning network component that determines respective weights assigned to respective node inputs of the learning network in accordance with a learning phase of the learning network and trains a variable separator component to differentially change learning rates of the learning network component. A differential rate component applies at least one update learning rate to adjust at least one weight assigned to at least one of the respective node inputs and applies at least one other update learning rate to adjust the respective weight assigned to at least one other of the respective node inputs in accordance with the variable separator component during the learning phase of the learning network. A differential rate component applies at least one update rate to adjust at least one weight assigned to at least one of the respective node inputs and applies at least one other update rate to adjust the respective weight assigned to at least one other of the respective node inputs in accordance with the learning phase of the learning network.

    IMAGE HARMONIZATION FOR DEEP LEARNING MODEL OPTIMIZATION
    35.
    发明申请
    IMAGE HARMONIZATION FOR DEEP LEARNING MODEL OPTIMIZATION 有权

    公开(公告)号：US20210334598A1

    公开(公告)日：2021-10-28

    申请号：US16858862

    申请日：2020-04-27

    Applicant: GE Precision Healthcare LLC

    Inventor： Tao Tan ,  Pál Tegzes ,  Levente Imre Török ,  Lehel Ferenczi ,  Gopal B. Avinash ,  László Ruskó ,  Gireesha Chinthamani Rao ,  Khaled Younis ,  Soumya Ghose

    IPC: G06K9/62

    Abstract: Techniques are described for optimizing deep learning model performance using image harmonization as a pre-processing step. According to an embodiment, a method comprises decomposing, by a system operatively coupled to a processor, an input image into sub-images. The method further comprises harmonizing the sub-images with corresponding reference sub-images of at least one reference image based on two or more different statistical values respectively calculated for the sub-images and the corresponding reference-sub images, resulting in transformation of the sub-images into modified sub-images images. In some implementations, the modified sub-images can be combined into a harmonized image having a more similar appearance to the at least one reference image relative to the input image. In other implementations, harmonized images and/or modified sub-images generated using these techniques can be used as ground-truth training samples for training one or more deep learning model to transform input images with appearance variations into harmonized images.

    ANNOTATION PIPELINE FOR MACHINE LEARNING ALGORITHM TRAINING AND OPTIMIZATION
    36.
    发明申请
    ANNOTATION PIPELINE FOR MACHINE LEARNING ALGORITHM TRAINING AND OPTIMIZATION 有权

    公开(公告)号：US20210034920A1

    公开(公告)日：2021-02-04

    申请号：US16527965

    申请日：2019-07-31

    Applicant: GE Precision Healthcare LLC

    Inventor： Marc T. Edgar ,  Travis R. Frosch ,  Gopal B. Avinash ,  Garry M. Whitley

    IPC: G06K9/62 ,  G16H50/20 ,  G06N5/04 ,  G06F17/24

    Abstract: Techniques are provided for enhancing the efficiency and accuracy of annotating data samples for supervised machine learning algorithms using an advanced annotation pipeline. According to an embodiment, a method can comprise collecting, by a system comprising a processor, unannotated data samples for input to a machine learning model and storing the unannotated data samples in an annotation queue. The method further comprises determining, by the system, annotation priority levels for respective unannotated data samples of the unannotated data samples, selecting, by the system from amongst different annotation techniques, one or more of the different annotation techniques for annotating the respective unannotated data samples based the annotation priority levels associated with the respective unannotated data samples.

    System and Method for Projection Enhancement for Synthetic 2D Image Generation
    38.
    发明公开
    System and Method for Projection Enhancement for Synthetic 2D Image Generation 审中-公开

    公开(公告)号：US20240193763A1

    公开(公告)日：2024-06-13

    申请号：US18077575

    申请日：2022-12-08

    Applicant: GE Precision Healthcare LLC

    Inventor： Sylvain Bernard ,  Dejun Wang ,  Buer Qi ,  Gopal B. Avinash ,  Gireesha Rao ,  Vincent Bismuth

    IPC: G06T7/00 ,  G06T5/00 ,  G06T7/194

    CPC classification number: G06T7/0012 ,  G06T5/001 ,  G06T7/194 ,  G06T2207/10116 ,  G06T2207/30096

    Abstract: Various methods and systems are provided for enhancing the generation of a synthetic 2D image from tomosynthesis projection images, such as a synthetic 2D image. To enhance the image, the image processing system utilizes a selected height interval to scan for objects of interest within a volume reconstructed from the tomosynthesis projection images. The height interval is larger than normal slices formed from the reconstructed volume, such that pixel information on larger masses can be obtained from adjacent slices within the volume. Further, the illustration of the object of interest in the synthetic 2D image can be modified by contributing pixel information from all tomosynthesis projections for the presentation of the object or interest. The use of pixel information from all tomosynthesis projections enhances the illustration of the high frequency components and the low frequency components of the object of interest within the enhanced image.

    Active surveillance and learning for machine learning model authoring and deployment
    39.
    发明授权
    Active surveillance and learning for machine learning model authoring and deployment 有权

    公开(公告)号：US11954610B2

    公开(公告)日：2024-04-09

    申请号：US16944762

    申请日：2020-07-31

    Applicant: GE Precision Healthcare LLC

    Inventor： Junpyo Hong ,  Venkata Ratnam Saripalli ,  Gopal B. Avinash ,  Karley Marty Yoder ,  Keith Bigelow

    IPC: G06N5/04 ,  G06N20/00

    CPC classification number: G06N5/04 ,  G06N20/00

    Abstract: Techniques are described for performing active surveillance and learning for machine learning (ML) model authoring and deployment workflows. In an embodiment, a method comprises applying, by a system comprising a processor, a primary ML model trained on a training dataset to data samples excluded from the training dataset to generate inferences based on the data samples. The method further comprises employing, by the system, one or more active surveillance techniques to regulate performance of the primary ML model in association with the applying, wherein the one or more active surveillance techniques comprise at least one of, performing a model scope evaluation of the primary ML model relative to the data samples or using a domain adapted version of the primary ML model to generate the inferences.

Patent Agency Ranking