Patents by Inventor Haofu Liao

Haofu Liao 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).

  • Patent number: 12171614
    Abstract: For three-dimensional segmentation from two-dimensional intracardiac echocardiography imaging, the three-dimension segmentation is output by a machine-learnt multi-task generator. Rather than the brute force approach of training the generator from 2D ICE images to output a 2D segmentation, the generator is trained from 3D information, such as a sparse ICE volume assembled from the 2D ICE images. Where sufficient ground truth data is not available, computed tomography or magnetic resonance data may be used as the ground truth for the sample sparse ICE volumes. The generator is trained to output both the 3D segmentation and a complete volume (i.e., more voxels represented than in the sparse ICE volume). The 3D segmentation may be further used to project to 2D as an input with an ICE image to another network trained to output a 2D segmentation for the ICE image. Display of the 3D segmentation and/or 2D segmentation may guide ablation of tissue in the patient.
    Type: Grant
    Filed: September 29, 2023
    Date of Patent: December 24, 2024
    Assignee: Siemens Medical Solutions USA, Inc.
    Inventors: Gareth Funka-Lea, Haofu Liao, Shaohua Kevin Zhou, Yefeng Zheng, Yucheng Tang
  • Publication number: 20240023927
    Abstract: For three-dimensional segmentation from two-dimensional intracardiac echocardiography imaging, the three-dimension segmentation is output by a machine-learnt multi-task generator. Rather than the brute force approach of training the generator from 2D ICE images to output a 2D segmentation, the generator is trained from 3D information, such as a sparse ICE volume assembled from the 2D ICE images. Where sufficient ground truth data is not available, computed tomography or magnetic resonance data may be used as the ground truth for the sample sparse ICE volumes. The generator is trained to output both the 3D segmentation and a complete volume (i.e., more voxels represented than in the sparse ICE volume). The 3D segmentation may be further used to project to 2D as an input with an ICE image to another network trained to output a 2D segmentation for the ICE image. Display of the 3D segmentation and/or 2D segmentation may guide ablation of tissue in the patient.
    Type: Application
    Filed: September 29, 2023
    Publication date: January 25, 2024
    Inventors: Gareth Funka-Lea, Haofu Liao, Shaohua Kevin Zhou, Yefeng Zheng, Yucheng Tang
  • Patent number: 11806189
    Abstract: For three-dimensional segmentation from two-dimensional intracardiac echocardiography imaging, the three-dimension segmentation is output by a machine-learnt multi-task generator. Rather than the brute force approach of training the generator from 2D ICE images to output a 2D segmentation, the generator is trained from 3D information, such as a sparse ICE volume assembled from the 2D ICE images. Where sufficient ground truth data is not available, computed tomography or magnetic resonance data may be used as the ground truth for the sample sparse ICE volumes. The generator is trained to output both the 3D segmentation and a complete volume (i.e., more voxels represented than in the sparse ICE volume). The 3D segmentation may be further used to project to 2D as an input with an ICE image to another network trained to output a 2D segmentation for the ICE image. Display of the 3D segmentation and/or 2D segmentation may guide ablation of tissue in the patient.
    Type: Grant
    Filed: November 28, 2022
    Date of Patent: November 7, 2023
    Assignee: Siemens Medical Solutions USA, Inc.
    Inventors: Gareth Funka-Lea, Haofu Liao, Shaohua Kevin Zhou, Yefeng Zheng, Yucheng Tang
  • Publication number: 20230113154
    Abstract: For three-dimensional segmentation from two-dimensional intracardiac echocardiography imaging, the three-dimension segmentation is output by a machine-learnt multi-task generator. Rather than the brute force approach of training the generator from 2D ICE images to output a 2D segmentation, the generator is trained from 3D information, such as a sparse ICE volume assembled from the 2D ICE images. Where sufficient ground truth data is not available, computed tomography or magnetic resonance data may be used as the ground truth for the sample sparse ICE volumes. The generator is trained to output both the 3D segmentation and a complete volume (i.e., more voxels represented than in the sparse ICE volume). The 3D segmentation may be further used to project to 2D as an input with an ICE image to another network trained to output a 2D segmentation for the ICE image. Display of the 3D segmentation and/or 2D segmentation may guide ablation of tissue in the patient.
    Type: Application
    Filed: November 28, 2022
    Publication date: April 13, 2023
    Inventors: Gareth Funka-Lea, Haofu Liao, Shaohua Kevin Zhou, Yefeng Zheng, Yucheng Tang
  • Patent number: 11534136
    Abstract: For three-dimensional segmentation from two-dimensional intracardiac echocardiography imaging, the three-dimension segmentation is output by a machine-learnt multi-task generator. The machine-learnt multi-task generator is trained from 3D information, such as a sparse ICE volume assembled from the 2D ICE images. The machine-learnt multi-task generator is trained to output both the 3D segmentation and a complete volume. The 3D segmentation may be used to project to 2D as an input with an ICE image to another network trained to output a 2D segmentation for the ICE image. Display of the 3D segmentation and/or 2D segmentation may guide ablation of tissue in the patient.
    Type: Grant
    Filed: September 13, 2018
    Date of Patent: December 27, 2022
    Assignee: Siemens Medical Solutions USA, Inc.
    Inventors: Gareth Funka-Lea, Haofu Liao, Shaohua Kevin Zhou, Yefeng Zheng, Yucheng Tang
  • Patent number: 11120582
    Abstract: An apparatus and method for coupled medical image formation and medical image signal recovery using a dual domain network is disclosed. The dual-domain network includes a first deep neural network (DNN) to perform signal recovery in a sensor signal domain and a second DNN to perform signal recovery in an image domain. A sensor signal is acquired by a sensor of a medical imaging device. A refined sensor signal is generated from the received sensor signal using the first DNN. A first reconstructed medical image is generated from the received sensor signal. A second reconstructed medical image is generated from the refined sensor signal generated by the first DNN. An enhanced medical image is generated based on the both the first reconstructed medical image and the second reconstructed medical image using the second DNN. The enhanced medical image generated by the second DNN is displayed.
    Type: Grant
    Filed: July 31, 2019
    Date of Patent: September 14, 2021
    Assignee: Z2SKY TECHNOLOGIES INC.
    Inventors: Shaohua Kevin Zhou, Haofu Liao, Wei-An Lin
  • Publication number: 20210035338
    Abstract: An apparatus and method for coupled medical image formation and medical image signal recovery using a dual domain network is disclosed. The dual-domain network includes a first deep neural network (DNN) to perform signal recovery in a sensor signal domain and a second DNN to perform signal recovery in an image domain. A sensor signal is acquired by a sensor of a medical imaging device. A refined sensor signal is generated from the received sensor signal using the first DNN. A first reconstructed medical image is generated from the received sensor signal. A second reconstructed medical image is generated from the refined sensor signal generated by the first DNN. An enhanced medical image is generated based on the both the first reconstructed medical image and the second reconstructed medical image using the second DNN. The enhanced medical image generated by the second DNN is displayed.
    Type: Application
    Filed: July 31, 2019
    Publication date: February 4, 2021
    Applicant: Z2Sky Technologies Inc.
    Inventors: Shaohua Kevin Zhou, Haofu Liao, Wei-An Lin
  • Patent number: 10709394
    Abstract: A method and apparatus for automated reconstruction of a 3D computed tomography (CT) volume from a small number of X-ray images is disclosed. A sparse 3D volume is generated from a small number of x-ray images using a tomographic reconstruction algorithm. A final reconstructed 3D CT volume is generated from the sparse 3D volume using a trained deep neural network. A 3D segmentation mask can also be generated from the sparse 3D volume using the trained deep neural network.
    Type: Grant
    Filed: January 15, 2018
    Date of Patent: July 14, 2020
    Assignee: Siemens Healthcare GmbH
    Inventors: Shaohua Kevin Zhou, Sri Venkata Anirudh Nanduri, Jin-hyeong Park, Haofu Liao
  • Publication number: 20190261945
    Abstract: For three-dimensional segmentation from two-dimensional intracardiac echocardiography imaging, the three-dimension segmentation is output by a machine-learnt multi-task generator. Rather than the brute force approach of training the generator from 2D ICE images to output a 2D segmentation, the generator is trained from 3D information, such as a sparse ICE volume assembled from the 2D ICE images. Where sufficient ground truth data is not available, computed tomography or magnetic resonance data may be used as the ground truth for the sample sparse ICE volumes. The generator is trained to output both the 3D segmentation and a complete volume (i.e., more voxels represented than in the sparse ICE volume). The 3D segmentation may be further used to project to 2D as an input with an ICE image to another network trained to output a 2D segmentation for the ICE image. Display of the 3D segmentation and/or 2D segmentation may guide ablation of tissue in the patient.
    Type: Application
    Filed: September 13, 2018
    Publication date: August 29, 2019
    Inventors: Gareth Funka-Lea, Haofu Liao, Shaohua Kevin Zhou, Yefeng Zheng, Yucheng Tang
  • Publication number: 20190216409
    Abstract: A method and apparatus for automated reconstruction of a 3D computed tomography (CT) volume from a small number of X-ray images is disclosed. A sparse 3D volume is generated from a small number of x-ray images using a tomographic reconstruction algorithm. A final reconstructed 3D CT volume is generated from the sparse 3D volume using a trained deep neural network. A 3D segmentation mask can also be generated from the sparse 3D volume using the trained deep neural network.
    Type: Application
    Filed: January 15, 2018
    Publication date: July 18, 2019
    Inventors: Shaohua Kevin Zhou, Sri Venkata Anirudh Nanduri, Jin-hyeong Park, Haofu Liao