-
公开(公告)号：US12045958B2
公开(公告)日：2024-07-23
申请号：US17378448
申请日：2021-07-16
发明人： Xiao Chen , Shuo Han , Zhang Chen , Shanhui Sun , Terrence Chen
IPC分类号： G06T5/70 , G01R33/48 , G01R33/565 , G06T7/00
CPC分类号： G06T5/70 , G01R33/4818 , G01R33/56509 , G06T7/0014 , G06T2207/10088 , G06T2207/20081 , G06T2207/20084
摘要： Neural network based systems, methods, and instrumentalities may be used to remove motion artifacts from magnetic resonance (MR) images. Such a neural network based system may be trained to perform the motion artifact removal tasks without reference (e.g., without using paired motion-contaminated and motion-free MR images). Various training techniques are described herein including one that feeds the neural network with pairs of MR images with different levels of motion contamination and forces the neural network learn to correct the motion contamination by transforming a first image of a contaminated pair into a second image of the contaminated pair. Other neural network training techniques are also described with an aim to reduce the reliance on training data that is difficult to obtain.
-
公开(公告)号：US20240233419A9
公开(公告)日：2024-07-11
申请号：US18128290
申请日：2023-03-30
发明人： Meng Zheng , Wenzhe Cui , Ziyan Wu , Arun Innanje , Benjamin Planche , Terrence Chen
CPC分类号： G06V20/70 , G06V10/235
摘要： Described herein are systems, methods, and instrumentalities associated with automatically annotating a 3D image dataset. The 3D automatic annotation may be accomplished based on a 2D manual annotation provided by an annotator and by propagating, using a set of machine-learning (ML) based techniques, the 2D manual annotation through sequences of 2D images associated with the 3D image dataset. The automatically annotated 3D image dataset may then be used to annotate other 3D image datasets upon passing a readiness assessment conducted using another set of ML based techniques. The automatic annotation of the images may be performed progressively, e.g., by processing a subset or batch of images at a time, and the ML based techniques may be trained to ensure consistency between a forward propagation and a backward propagation.
-
公开(公告)号：US20240169486A1
公开(公告)日：2024-05-23
申请号：US17989205
申请日：2022-11-17
发明人： Yikang Liu , Zhang Chen , Xiao Chen , Shanhui Sun , Terrence Chen
CPC分类号： G06T5/50 , G06T5/002 , G06T5/003 , G06T2207/10016 , G06T2207/10121 , G06T2207/20081 , G06T2207/20084 , G06T2207/30004
摘要： Deblurring and denoising a medical image such as X-ray fluoroscopy images may be challenging, and deep-learning based techniques may be employed to meet the challenge. An artificial neural network (ANN) may be trained using training images with synthetic noise and as well as training images with real noise. The parameters of the ANN may be adjusted during the training based on at least a first loss designed to maintain continuity between consecutive medical images generated by the ANN and a second loss designed to maintain similarity of patches inside a medical image generated by the ANN. The parameters of the ANN may be further adjusted based on a third loss that may be calculated from ground truth associated with the synthetic training images. Transfer learning between the synthetic images and the real images may be accomplished using these techniques.
-
公开(公告)号：US20240153094A1
公开(公告)日：2024-05-09
申请号：US17981988
申请日：2022-11-07
发明人： Yikang Liu , Shanhui Sun , Terrence Chen
CPC分类号： G06T7/11 , G06T7/0012 , G16H50/50 , G06T2207/30101
摘要： Described herein are systems, methods, and instrumentalities associated with automatically annotating a tubular structure (e.g., such as a blood vessel, a catheter, etc.) in medical images. The automatic annotation may be accomplished using a machine-learning image annotation model and based on a marking of the tubular structure created or confirmed by a user. A user interface may be provided for a user to create, modify, and/or confirm the marking, and the ML model may be trained using a training dataset that comprises marked images of the tubular structure paired with ground truth annotations of the tubular structure.
-
公开(公告)号：US11967004B2
公开(公告)日：2024-04-23
申请号：US17378465
申请日：2021-07-16
发明人： Zhang Chen , Shanhui Sun , Xiao Chen , Terrence Chen
IPC分类号： G06T11/00 , A61B5/00 , A61B5/055 , G06F18/214 , G06F18/22 , G06K9/62 , G06N3/04 , G06N3/08 , G06T5/50
CPC分类号： G06T11/005 , A61B5/055 , A61B5/7267 , G06F18/214 , G06F18/22 , G06N3/04 , G06N3/08 , G06T5/50 , G06T2207/10088 , G06T2207/20081 , G06T2207/20084 , G06T2207/20221
摘要： Disclosed herein are systems, methods, and instrumentalities associated with reconstructing magnetic resonance (MR) images based on under-sampled MR data. The MR data include 2D or 3D information, and may encompass multiple contrasts and multiple coils. The MR images are reconstructed using deep learning (DL) methods, which may accelerate the scan and/or image generation process. Challenges imposed by the large quantity of the MR data and hardware limitations are overcome by separately reconstructing MR images based on respective subsets of contrasts, coils, and/or readout segments, and then combining the reconstructed MR images to obtain desired multi-contrast results.
-
公开(公告)号：US11948288B2
公开(公告)日：2024-04-02
申请号：US17340635
申请日：2021-06-07
发明人： Xiao Chen , Shuo Han , Zhang Chen , Shanhui Sun , Terrence Chen
CPC分类号： G06T5/70 , G06N3/08 , G16H30/20 , G16H30/40 , G16H50/20 , G16H50/70 , G06T2207/10088 , G06T2207/20056 , G06T2207/20081 , G06T2207/20084 , G16H50/50
摘要： Motion contaminated magnetic resonance (MR) images for training an artificial neural network to remove motion artifacts from the MR images are difficult to obtain. Described herein are systems, methods, and instrumentalities for injecting motion artifacts into clean MR images and using the artificially contaminated images for machine learning and neural network training. The motion contaminated MR images may be created based on clean source MR images that are associated with multiple physiological cycles of a scanned object, and by deriving MR data segments for the multiple physiological cycles based on the source MR images. The MR data segments thus derived may be combined to obtain a simulated MR data set, from which one or more target MR images may be generated to exhibit a motion artifact. The motion artifact may be created by manipulating the source MR images and/or controlling the manner in which the MR data set or the target MR images are generated.
-
公开(公告)号：US20240099774A1
公开(公告)日：2024-03-28
申请号：US17955279
申请日：2022-09-28
发明人： Meng Zheng , Benjamin Planche , Ziyan Wu , Terrence Chen
CPC分类号： A61B34/10 , A61B34/30 , G06T17/20 , G16H50/50 , A61B2034/105 , A61B2034/107
摘要： Systems, methods and instrumentalities are described herein for automatically devising and executing a surgical plan associated with a patient in a medical environment, e.g., under the supervision of a medical professional. The surgical plan may be devised based on images of the medical environment captured by one or more sensing devices. A processing device may determine, based on all or a first subset of the images, a patient model that may indicate a location and a shape of an anatomical structure of the patient and determine, based on all or a second subset of the images, an environment model that may indicate a three-dimensional (3D) spatial layout of the medical environment. The surgical plan may be devised based on the patient model and the environment model, and may indicate at least a movement path of a medical device towards the anatomical structure of the patient.
-
公开(公告)号：US20240079128A1
公开(公告)日：2024-03-07
申请号：US17901349
申请日：2022-09-01
发明人： Terrence Chen
摘要： Traditional ways of seeking and receiving healthcare services are time-consuming and cumbersome. A digital healthcare service platform built on artificial intelligence (AI) technologies may improve the experience and efficiency associated with these services. The digital healthcare platform may use AI models trained for image classification and/or natural language processing to generate preliminary diagnoses for a care seeker based on images or descriptions provided by the care seeker. The digital healthcare platform may also use AI models to match service providers with the care seeker, and/or manage the logistical aspects of a service (e.g., coordinating activities, scheduling appointments, etc.) for the care seeker.
-
公开(公告)号：US20240062047A1
公开(公告)日：2024-02-22
申请号：US17891702
申请日：2022-08-19
发明人： Zhang Chen , Siyuan Dong , Shanhui Sun , Xiao Chen , Yikang Liu , Terrence Chen
IPC分类号： G06N3/04
CPC分类号： G06N3/0472 , G06N3/0454 , G06T2207/20081 , G06T2207/20084
摘要： Deep learning-based systems, methods, and instrumentalities are described herein for MRI reconstruction and/or refinement. An MRI image may be reconstructed based on under-sampled MRI information and a generative model may be trained to refine the reconstructed image, for example, by increasing the sharpness of the MRI image without introducing artifacts into the image. The generative model may be implemented using various types of artificial neural networks including a generative adversarial network. The model may be trained based on an adversarial loss and a pixel-wise image loss, and once trained, the model may be used to improve the quality of a wide range of 2D or 3D MRI images including those of a knee, brain, or heart.
-
10.发明公开公开(公告)号：US20230419740A1
公开(公告)日：2023-12-28
申请号：US17851190
申请日：2022-06-28
发明人： Benjamin Planche , Ziyan Wu , Meng Zheng , Terrence Chen
CPC分类号： G06V40/70 , G06T7/0012 , G06V40/15 , G06T2207/10088 , G06T2207/10104 , G06T2207/10108 , G06T2207/10116 , G06T2207/10081
摘要： A non-invasive biometric system includes a processor that is configured to control a scanner, which is configured to scan and capture one or more anatomical images of a body of a target person. The processor is further configured to identify one or more anatomical structures in the captured one or more anatomical images and extract anatomical features for the identified one or more anatomical structures. The processor is further configured to register the extracted anatomical features for the identified one or more identified anatomical structures to a posture and an external appearance of the target person. The processor is further configured to encode and utilize the extracted anatomical features as biometric data, which is unique for the target person, and may be used for authentication of the target person.
-
-
-
-
-
-
-
-
-
搜索结果
110 条记录 (耗时 0.033 秒)
