Abstract: An illustrative room identification system performs augmented reality room identification based on free-hand image capture and room-object profile data. The room identification system obtains a set of reference room-object profiles for a set of rooms associated with an experience presented by an augmented reality presentation device. The room identification system generates a live room-object profile based on an image captured by the presentation device during the augmented reality experience, the image depicting a particular room of the set of rooms. Based on the live room-object profile and the set of reference room-object profiles, the room identification system determines that the presentation device is located in the particular room of the set of rooms and provides, to the presentation device, an indication of the determination that the presentation device is located in the particular room. Corresponding methods and systems are also disclosed.

Abstract: A mixed reality (MR) system and method performs alignment of a digital twin and the corresponding real-world object using 3D deep neural network structures using multimodal fusion and simplified machine learning to cluster label distributions (output of 3D deep neural network trained by generic 3D benchmark dataset) that are used to reduce the training data requirements to directly train a 3D deep neural network structures. In one embodiment, multiple 3D deep neural network structures, such as PointCNN, 3D-Bonet, RandLA, etc., may be trained by different generic 3D benchmark datasets, such as ScanNet, ShapeNet, S3DIS, inadequate 3D training dataset, etc.

Abstract: A computer-implemented method to improve the point collection process during registration of a bone for a computer-assisted surgical procedure is provided. Based on bone digitization data, a simulation is performed to confirm the accuracy of the registration for different digitization regions. Results are tested to identify which digitization regions meet a predefined accuracy requirement. The resulting information is used to perform a computer-assisted surgical procedure. A computerized simulation method for registration of a bone for a computer-assisted surgical procedure is also provided based on processor executing random stroking an expected exposed surface of a bone model with multiple of stroke curves to cover most of the bone model surface with uniform noise and a random sample consensus is applied to remove outlying point to yield the best registration results, to find the top subset as to overlap. A method to perform computer-assisted surgery is also provided.

Abstract: A computer-implemented method to improve the point collection process during registration of a bone for a computer-assisted surgical procedure is provided. Based on bone digitization data, a simulation is performed to confirm the accuracy of the registration for different digitization regions. Results are tested to identify which digitization regions meet a predefined accuracy requirement. The resulting information is used to perform a computer-assisted surgical procedure. A computerized simulation method for registration of a bone for a computer-assisted surgical procedure is also provided based on processor executing random stroking an expected exposed surface of a bone model with multiple of stroke curves to cover most of the bone model surface with uniform noise and a random sample consensus is applied to remove outlying point to yield the best registration results, to find the top subset as to overlap. A method to perform computer-assisted surgery is also provided.

Abstract: An illustrative room identification system performs augmented reality room identification based on free-hand image capture and room-object profile data. The room identification system obtains a set of reference room-object profiles for a set of rooms associated with an experience presented by an augmented reality presentation device. The room identification system generates a live room-object profile based on an image captured by the presentation device during the augmented reality experience, the image depicting a particular room of the set of rooms. Based on the live room-object profile and the set of reference room-object profiles, the room identification system determines that the presentation device is located in the particular room of the set of rooms and provides, to the presentation device, an indication of the determination that the presentation device is located in the particular room. Corresponding methods and systems are also disclosed.

Abstract: A mixed reality (MR) system and method performs three dimensional (3D) tracking using 3D deep neural network structures in which multimodal fusion and simplified machine learning to only cluster label distribution (output of 3D deep neural network trained by generic 3D benchmark dataset) is used to reduce the training data requirements of to directly train a 3D deep neural network structures for non-generic user case. In one embodiment, multiple 3D deep neural network structures, such as PointCNN, 3D-Bonet, RandLA, etc., may be trained by different generic 3D benchmark datasets, such as ScanNet, ShapeNet, S3DIS, inadequate 3D training dataset, etc.

Abstract: In an example, a method includes selecting a user-selected seed voxel from a medical image. The user-selected seed voxel is shifted to an adjusted seed voxel in an evaluation region based on an evaluation. The evaluation region includes the user-selected seed voxel and at least one other voxel within a predetermined spatial distance. A given region in the image volume is filled with voxels to provide a segmented volume for the given region based on the adjusted seed voxel and segmentation filter parameters. A segmentation region is evaluated to ascertain a quality of the fill. Seed candidates may be determined from a set of the filled voxels adjacent a boundary of the for another region given region in response to determining that the quality of the fill indicates a valid segmentation for the anatomic vessel structure in the given region. Segmented vessel image data is stored.

Abstract: A computer-implemented method to improve the point collection process during registration of a bone for a computer-assisted surgical procedure is provided. Based on bone digitization data, a simulation is performed to confirm the accuracy of the registration for different digitization regions. Results are tested to identify which digitization regions meet a predefined accuracy requirement. The resulting information is used to perform a computer-assisted surgical procedure. A computerized simulation method for registration of a bone for a computer-assisted surgical procedure is also provided based on processor executing random stroking an expected exposed surface of a bone model with multiple of stroke curves to cover most of the bone model surface with uniform noise and a random sample consensus is applied to remove outlying point to yield the best registration results, to find the top subset as to overlap. A method to perform computer-assisted surgery is also provided.

Abstract: In an example, a method includes selecting a user-selected seed voxel from a medical image. The user-selected seed voxel is shifted to an adjusted seed voxel in an evaluation region based on an evaluation. The evaluation region includes the user-selected seed voxel and at least one other voxel within a predetermined spatial distance. A given region in the image volume is filled with voxels to provide a segmented volume for the given region based on the adjusted seed voxel and segmentation filter parameters. A segmentation region is evaluated to ascertain a quality of the fill. Seed candidates may be determined from a set of the filled voxels adjacent a boundary of the for another region given region in response to determining that the quality of the fill indicates a valid segmentation for the anatomic vessel structure in the given region. Segmented vessel image data is stored.