-
公开(公告)号：US12014814B2
公开(公告)日：2024-06-18
申请号：US16775180
申请日：2020-01-28
Applicant: GE Precision Healthcare LLC
Inventor： Katelyn Nye , Gopal Avinash , Pal Tegzes , Gireesha Rao
CPC classification number: G16H30/40 , G06F18/217 , G06F18/285 , G06F18/40 , G06N5/02 , G06N20/00 , G06T7/0012 , G16H50/50 , G06T2207/10116 , G06V2201/03
Abstract: Methods and systems are provided for tuning a static model with multiple operating points to adjust model performance without retraining the model or triggering a new regulatory clearance. In one embodiment, a method comprises, responsive to a request to tune a model, obtaining a tuning dataset including a set of medical images, executing the model using the set of medical images as input to generate model tuning output, and determining, for each operating point of a set of operating points, a set of tuning metric values based on the tuning dataset and the model tuning output relative to each operating point. An operating point from the set of operating points may be selected based on each set of tuning metric values and, upon a request to analyze a subsequent medical image, a representation of a finding output from the static model executed at the selected operating point.
-
公开(公告)号：US11842485B2
公开(公告)日：2023-12-12
申请号：US17192804
申请日：2021-03-04
Applicant: GE Precision Healthcare LLC
Inventor： Tao Tan , Máté Fejes , Gopal Avinash , Ravi Soni , Bipul Das , Rakesh Mullick , Pál Tegzes , Lehel Ferenczi , Vikram Melapudi , Krishna Seetharam Shriram
CPC classification number: G06T7/0012 , G06N3/08 , G06T15/08 , G06T2207/10088 , G06T2207/10104
Abstract: Methods and systems are provided for inferring thickness and volume of one or more object classes of interest in two-dimensional (2D) medical images, using deep neural networks. In an exemplary embodiment, a thickness of an object class of interest may be inferred by acquiring a 2D medical image, extracting features from the 2D medical image, mapping the features to a segmentation mask for an object class of interest using a first convolutional neural network (CNN), mapping the features to a thickness mask for the object class of interest using a second CNN, wherein the thickness mask indicates a thickness of the object class of interest at each pixel of a plurality of pixels of the 2D medical image; and determining a volume of the object class of interest based on the thickness mask and the segmentation mask.
-
公开(公告)号：US20240296937A1
公开(公告)日：2024-09-05
申请号：US18662869
申请日：2024-05-13
Applicant: GE Precision Healthcare LLC
Inventor： Katelyn Nye , Gopal Avinash , Pal Tegzes , Gireesha Rao
CPC classification number: G16H30/40 , G06F18/217 , G06F18/285 , G06F18/40 , G06N5/02 , G06N20/00 , G06T7/0012 , G16H50/50 , G06T2207/10116 , G06V2201/03
Abstract: Methods and systems are provided for tuning a static model with multiple operating points to adjust model performance without retraining the model or triggering a new regulatory clearance. In one embodiment, a method comprises, responsive to a request to tune a model, obtaining a tuning dataset including a set of medical images, executing the model using the set of medical images as input to generate model tuning output, and determining, for each operating point of a set of operating points, a set of tuning metric values based on the tuning dataset and the model tuning output relative to each operating point. An operating point from the set of operating points may be selected based on each set of tuning metric values and, upon a request to analyze a subsequent medical image, a representation of a finding output from the static model executed at the selected operating point.
-
Patent Agency Ranking
公开(公告)号：US20240164845A1
公开(公告)日：2024-05-23
申请号：US18385448
申请日：2023-10-31
Applicant: GE Precision Healthcare LLC
Inventor： Pál Tegzes , Zita Herczeg , Hongxu Yang , Zoltán Kiss , Balázs P. Cziria , Poonam Dalal , Alec Baenen , Gireesha Rao , Beth Heckel , Pulak Goswami , Dennis Zhou , Gopal Avinash , Lehel Ferenczi , Katelyn Nye , Noah Thompson Orfield , Emma Neal , Sarah Ouadah
CPC classification number: A61B34/20 , A61B6/12 , A61B6/463 , A61B6/54 , G06T7/74 , G06T11/00 , A61B2034/2065 , G06T2207/30012 , G06T2207/30204 , G06T2210/41
Abstract: An image processing system and method is provided. The image processing system includes a display, a processor, and a memory. The memory stores processor-executable code that when executed by the processor causes receiving an image of a region of interest of a patient with a medical catheter, tube or line disposed within the region of interest, detecting the medical tube or line within the image, generating a patient coordinate system relative to an anatomy of the patient within the image, generating a combined image by superimposing a first graphical marker on the image that indicates an end of the medical catheter, tube or line, and a second graphical marker on the image that indicates patient coordinate system, and displaying the combined image on the display. In addition, the system assesses common visualizable complications associated with CVC placement, including but not limited to hydrothorax, pneumothorax, pneumomediastinum and CVC position changes between x-rays taken at different times.
-
公开(公告)号：US20240078669A1
公开(公告)日：2024-03-07
申请号：US18497912
申请日：2023-10-30
Applicant: GE Precision Healthcare LLC
Inventor： Tao Tan , Máté Fejes , Gopal Avinash , Ravi Soni , Bipul Das , Rakesh Mullick , Pál Tegzes , Lehel Ferenczi , Vikram Melapudi , Krishna Seetharam Shriram
CPC classification number: G06T7/0012 , G06N3/08 , G06T15/08 , G06T2207/10088 , G06T2207/10104
Abstract: Methods and systems are provided for inferring thickness and volume of one or more object classes of interest in two-dimensional (2D) medical images, using deep neural networks. In an exemplary embodiment, a thickness of an object class of interest may be inferred by acquiring a 2D medical image, extracting features from the 2D medical image, mapping the features to a segmentation mask for an object class of interest using a first convolutional neural network (CNN), mapping the features to a thickness mask for the object class of interest using a second CNN, wherein the thickness mask indicates a thickness of the object class of interest at each pixel of a plurality of pixels of the 2D medical image; and determining a volume of the object class of interest based on the thickness mask and the segmentation mask.
-
公开(公告)号：US20220284570A1
公开(公告)日：2022-09-08
申请号：US17192804
申请日：2021-03-04
Applicant: GE Precision Healthcare LLC
Inventor： Tao Tan , Máté Fejes , Gopal Avinash , Ravi Soni , Bipul Das , Rakesh Mullick , Pál Tegzes , Lehel Ferenczi , Vikram Melapudi , Krishna Seetharam Shriram
Abstract: Methods and systems are provided for inferring thickness and volume of one or more object classes of interest in two-dimensional (2D) medical images, using deep neural networks. In an exemplary embodiment, a thickness of an object class of interest may be inferred by acquiring a 2D medical image, extracting features from the 2D medical image, mapping the features to a segmentation mask for an object class of interest using a first convolutional neural network (CNN), mapping the features to a thickness mask for the object class of interest using a second CNN, wherein the thickness mask indicates a thickness of the object class of interest at each pixel of a plurality of pixels of the 2D medical image; and determining a volume of the object class of interest based on the thickness mask and the segmentation mask.
-
公开(公告)号：US20200337648A1
公开(公告)日：2020-10-29
申请号：US16697736
申请日：2019-11-27
Applicant: GE Precision Healthcare LLC
Inventor： Venkata Ratna Saripalli , Gopal Avinash , Min Zhang , Ravi Soni , Jiahui Guan , Dibyajyoti Pati , Zili Ma
Abstract: Systems, apparatus, instructions, and methods for medical machine time-series event data processing are disclosed. An example time series event data processing apparatus includes memory storing instructions and one-dimensional time series healthcare-related data; and at least one processor. The example at least one processor is to: execute artificial intelligence model(s) trained on aggregated time series data to at least one of a) predict a future medical machine event, b) detect a medical machine event, or c) classify the medical machine event using the one-dimensional time series healthcare-related data; when the artificial intelligence model(s) are executed to predict the future medical machine event, output an alert related to the predicted future medical machine event to trigger a next action; and when the artificial intelligence model(s) are executed to detect and/or classify the medical machine event, label the medical machine event and output the labeled event to trigger the next action.
-
公开(公告)号：US20210232967A1
公开(公告)日：2021-07-29
申请号：US16775180
申请日：2020-01-28
Applicant: GE Precision Healthcare LLC
Inventor： Katelyn Nye , Gopal Avinash , Pal Tegzes , Gireesha Rao
Abstract: Methods and systems are provided for tuning a static model with multiple operating points to adjust model performance without retraining the model or triggering a new regulatory clearance. In one embodiment, a method comprises, responsive to a request to tune a model, obtaining a tuning dataset including a set of medical images, executing the model using the set of medical images as input to generate model tuning output, and determining, for each operating point of a set of operating points, a set of tuning metric values based on the tuning dataset and the model tuning output relative to each operating point. An operating point from the set of operating points may be selected based on each set of tuning metric values and, upon a request to analyze a subsequent medical image, a representation of a finding output from the static model executed at the selected operating point.
-
-
-
-
-
-
-
Search Results
8
records
(took 0.014 seconds)
