FEATURE EXTRACTION VIA FEDERATED SELF-SUPERVISED LEARNING

    公开(公告)号:US20240303506A1

    公开(公告)日:2024-09-12

    申请号:US18598123

    申请日:2024-03-07

    CPC classification number: G06N3/098 G06N3/0895

    Abstract: One embodiment of the present invention sets forth a technique for training a machine learning model to perform feature extraction. The technique includes executing a student version of the machine learning model to generate a first set of features from a first set of image crops and executing a teacher version of the machine learning model to generate a second set of features from a second set of image crops. The technique also includes training the student version of the machine learning model based on one or more losses computed between the first and second sets of features. The technique further includes transmitting the trained student version of the machine learning model to a server, wherein the trained student version can be aggregated by the server with additional trained student versions of the machine learning model to generate a global version of the machine learning model.

    UNGROUNDED MASTER CONTROL DEVICES AND METHODS OF USE

    公开(公告)号:US20220313379A1

    公开(公告)日:2022-10-06

    申请号:US17833667

    申请日:2022-06-06

    Inventor: Anthony M. JARC

    Abstract: A mechanically ungrounded master control device comprises a rigid chassis for engagement with one or more fingers of a user's hand and at least one pivotable finger engagement device for controlling movement of a tool end effector. The control device also includes at least one sensor for detecting a position and orientation of the rigid chassis in a surgical environment and a switch coupled to the rigid chassis and manipulatable by a finger of the user's hand while another finger of the user's hand is engaged with the at least one pivotable finger engagement device.

    SIMULATOR SYSTEM FOR MEDICAL PROCEDURE TRAINING

    公开(公告)号:US20220415210A1

    公开(公告)日:2022-12-29

    申请号:US17902678

    申请日:2022-09-02

    Abstract: Implementations relate to medical simulations for medical procedure training. In some implementations, a system includes a simulation processing component including at least one processor and which generates a virtual environment using position signals that describe at least one of a position and a configuration of a physical surgical instrument relative to a physical surgical site. The simulation processing component updates the virtual environment according to changes in the position signals and according to control signals corresponding to inputs by a user of the system. The updating in cludes moving a virtual surgical instrument within the virtual environment, where an interaction of the virtual surgical instrument with a virtual surgical site is defined at least partly by a physical relationship between the physical surgical instrument and the physical surgical site. The simulation processing component outputs a representation of a simulation state signal indicative of a current state of the virtual environment.

    ANATOMICAL MODEL AND METHOD FOR SURGICAL TRAINING
    6.
    发明申请
    ANATOMICAL MODEL AND METHOD FOR SURGICAL TRAINING 审中-公开
    解剖学模型和手术训练方法

    公开(公告)号:US20140051049A1

    公开(公告)日:2014-02-20

    申请号:US13968253

    申请日:2013-08-15

    CPC classification number: G09B23/30 G09B23/285

    Abstract: Implementations relate to anatomical models and surgical training. In some implementations, an anatomical training model includes a base portion and a top portion that form a hollow space between the base portion and top portion. A plurality of holes are positioned in the top portion. The model includes a plurality of cannula supports, where each cannula support is aligned with one or more corresponding holes in the top portion.

    Abstract translation: 实施涉及解剖模型和手术训练。 在一些实施方案中,解剖训练模型包括在基部和顶部之间形成中空空间的基部和顶部。 多个孔位于顶部。 该模型包括多个插管支撑件,其中每个插管支架与顶部中的一个或多个对应的孔对齐。

Patent Agency Ranking