Systems and Methods for Surgical Training Model

    公开(公告)号:US20230136935A1

    公开(公告)日:2023-05-04

    申请号:US17787405

    申请日:2020-12-18

    摘要: Disclosed are a method for creating a surgical training model, a surgical training model apparatus, a bone model, an article that emulates tissue of an animal musculoskeletal system, an article that emulates animal fat tissue, and an article that emulates animal skin tissue. One version of the method comprises placing a spinal vertebrae model in a cavity model that emulates an animal body cavity; forming a first layer on top of the vertebrae model, wherein the first layer emulates an animal muscle tissue; placing a second layer over the first layer, wherein the second layer emulates an animal fat tissue; and placing a third layer over the second layer, wherein the third layer emulates an animal skin tissue. The spinal vertebrae model can be 3D printed from a thermoplastic polymer and infiltrated with a foam into an interior space of the 3D printed spinal vertebrae model.

    Systems, Methods, and Media for Training a Model for Improved Out of Distribution Performance

    公开(公告)号:US20230126226A1

    公开(公告)日:2023-04-27

    申请号:US17970771

    申请日:2022-10-21

    IPC分类号: G06V10/774 G06V10/776

    摘要: In accordance with some embodiments, systems, methods, and media for training a model for improved out of distribution performance are provided. In some embodiments, the method comprises: receiving a plurality of datasets, each associated with a different environment e; initializing data representation parameters associated with a model; providing the datasets as input to the model; receiving, from the model, an output associated with each input; determining an optimal classifier for the data representation parameters using an invariance penalty based on a square root of matrix e(φ):=EXe[(φ(Xe)φ(Xe)T] for e, where φ represents the data representation parameters, and φ(xe) is the dataset associated with environment e modified based on the data representation parameters; calculating a loss value for the optimal classifier across the datasets; and modifying the data representation parameters based on the loss value.