Patents by Inventor Shaswat Lenka

Shaswat Lenka has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • System and method for labeling medical data to generate labeled training data
    Patent number: 11763945
    Abstract: A method of automated labeling of medical data to generate labeled training data for training a learning algorithm is provided, wherein the medical data includes both machine-generated data elements and human-inputted data elements. The method includes using a knowledge graph to generate a first label for each machine-generated data element and each human-inputted data element. A set of trained classifier models are used to generate a second label for each machine-generated data element and each human-inputted data element. An unsupervised clustering model is used to generate a third label for each machine-generated data element and each human-inputted data element. A final label for each data element is then generated based on the first label, the second label, and the third label for the respective data element to output a labeled training dataset.
    Type: Grant
    Filed: December 16, 2019
    Date of Patent: September 19, 2023
    Assignee: GE Precision Healthcare LLC
    Inventors: Sankit Maroo, Jagan Selvaraj, Kiran Muthigi, Shaswat Lenka
  • System and Method for Labeling Medical Data to Generate Labeled Training Data
    Publication number: 20210182606
    Abstract: A method of automated labeling of medical data to generate labeled training data for training a learning algorithm is provided, wherein the medical data includes both machine-generated data elements and human-inputted data elements. The method includes using a knowledge graph to generate a first label for each machine-generated data element and each human-inputted data element. A set of trained classifier models are used to generate a second label for each machine-generated data element and each human-inputted data element. An unsupervised clustering model is used to generate a third label for each machine-generated data element and each human-inputted data element. A final label for each data element is then generated based on the first label, the second label, and the third label for the respective data element to output a labeled training dataset.
    Type: Application
    Filed: December 16, 2019
    Publication date: June 17, 2021
    Applicant: GE Precision Healthcare LLC
    Inventors: Sankit Maroo, Jagan Selvaraj, Kiran Muthigi, Shaswat Lenka