公开(公告)号：US12131446B2

公开(公告)日：2024-10-29

申请号：US17368534

申请日：2021-07-06

Applicant: GE Precision Healthcare LLC

Inventor： Rajesh Veera Venkata Lakshmi Langoju ,  Prasad Sudhakara Murthy ,  Utkarsh Agrawal ,  Bhushan D. Patil ,  Bipul Das

IPC: G06T5/73 ,  G06N20/00 ,  G06T3/4053 ,  G06T5/20

CPC classification number: G06T5/73 ,  G06N20/00 ,  G06T3/4053 ,  G06T5/20 ,  G06T2207/20081

Abstract: Systems/techniques that facilitate self-supervised deblurring are provided. In various embodiments, a system can access an input image generated by an imaging device. In various aspects, the system can train, in a self-supervised manner based on a point spread function of the imaging device, a machine learning model to deblur the input image. More specifically, the system can append to the model one or more non-trainable convolution layers having a blur kernel that is based on the point spread function of the imaging device. In various aspects, the system can feed the input image to the model, the model can generate a first output image based on the input image, the one or more non-trainable convolution layers can generate a second output image by convolving the first output image with the blur kernel, and the system can update parameters of the model based on a difference between the input image and the second output image.

公开(公告)号：US20230177747A1

公开(公告)日：2023-06-08

申请号：US17543234

申请日：2021-12-06

Applicant: GE Precision Healthcare LLC

Inventor： Rajesh Veera Venkata Lakshmi Langoju ,  Utkarsh Agrawal ,  Bipul Das ,  Risa Shigemasa ,  Yasuhiro Imai ,  Jiang Hsieh

IPC: G06T11/00 ,  G06T5/00 ,  G06T5/50 ,  G06N20/20

CPC classification number: G06T11/008 ,  G06N20/20 ,  G06T5/002 ,  G06T5/50

Abstract: Systems/techniques that facilitate machine learning generation of low-noise and high structural conspicuity images are provided. In various embodiments, a system can access an image and can apply at least one of image denoising or image resolution enhancement to the image, thereby yielding a first intermediary image. In various instances, the system can generate, via execution of a plurality of machine learning models, a plurality of second intermediary images based on the first intermediary image, wherein a given machine learning model in the plurality of machine learning models receives as input the first intermediary image, wherein the given machine learning model produces as output a given second intermediary image in the plurality of second intermediary images, and wherein the given second intermediary image represents a kernel-transformed version of the first intermediary image. In various cases, the system can generate a blended image based on the plurality of second intermediary images.

公开(公告)号：US12141900B2

公开(公告)日：2024-11-12

申请号：US17543234

申请日：2021-12-06

Applicant: GE Precision Healthcare LLC

Inventor： Rajesh Veera Venkata Lakshmi Langoju ,  Utkarsh Agrawal ,  Bipul Das ,  Risa Shigemasa ,  Yasuhiro Imai ,  Jiang Hsieh

IPC: G06T11/00 ,  G06N20/20 ,  G06T5/50 ,  G06T5/70

Abstract: Systems/techniques that facilitate machine learning generation of low-noise and high structural conspicuity images are provided. In various embodiments, a system can access an image and can apply at least one of image denoising or image resolution enhancement to the image, thereby yielding a first intermediary image. In various instances, the system can generate, via execution of a plurality of machine learning models, a plurality of second intermediary images based on the first intermediary image, wherein a given machine learning model in the plurality of machine learning models receives as input the first intermediary image, wherein the given machine learning model produces as output a given second intermediary image in the plurality of second intermediary images, and wherein the given second intermediary image represents a kernel-transformed version of the first intermediary image. In various cases, the system can generate a blended image based on the plurality of second intermediary images.

公开(公告)号：US20230052595A1

公开(公告)日：2023-02-16

申请号：US17403017

申请日：2021-08-16

Applicant: GE Precision Healthcare LLC

Inventor： Rajesh Veera Venkata Lakshmi Langoju ,  Utkarsh Agrawal ,  Bipul Das ,  Risa Shigemasa ,  Yasuhiro Imai ,  Jiang Hsieh

IPC: G06T5/00 ,  G06T3/40 ,  G06T5/50

Abstract: Techniques are described for enhancing the quality of three-dimensional (3D) anatomy scan images using deep learning. According to an embodiment, a system is provided that comprises a memory that stores computer executable components, and a processor that executes the computer executable components stored in the memory. The computer executable components comprise a reception component that receives a scan image generated from 3D scan data relative to a first axis of a 3D volume, and an enhancement component that applies an enhancement model to the scan image to generate an enhanced scan image having a higher resolution relative to the scan image. The enhancement model comprises a deep learning neural network model trained on training image pairs respectively comprising a low-resolution scan image and a corresponding high-resolution scan image respectively generated relative to a second axis of the 3D volume.

公开(公告)号：US20230013779A1

公开(公告)日：2023-01-19

申请号：US17368534

申请日：2021-07-06

Applicant: GE Precision Healthcare LLC

Inventor： Rajesh Veera Venkata Lakshmi Langoju ,  Prasad Sudhakara Murthy ,  Utkarsh Agrawal ,  Bhushan D. Patil ,  Bipul Das

IPC: G06T5/00 ,  G06T5/20 ,  G06T3/40 ,  G06N20/00

Abstract: Systems/techniques that facilitate self-supervised deblurring are provided. In various embodiments, a system can access an input image generated by an imaging device. In various aspects, the system can train, in a self-supervised manner based on a point spread function of the imaging device, a machine learning model to deblur the input image. More specifically, the system can append to the model one or more non-trainable convolution layers having a blur kernel that is based on the point spread function of the imaging device. In various aspects, the system can feed the input image to the model, the model can generate a first output image based on the input image, the one or more non-trainable convolution layers can generate a second output image by convolving the first output image with the blur kernel, and the system can update parameters of the model based on a difference between the input image and the second output image.