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.

Various examples are provided related to identification of anthropometric information for fitting of compression garments. In one example, a method of generating compression garment fit information includes acquiring images including a selected body part or body area in need of compression therapy; processing the acquired images along with a library of compression garment fit information; generating a 3D reconstruction of the selected body part/area; deriving anthropometric information for the selected body part/area from the 3D reconstruction; providing a compression value corresponding to a prescribed or intended amount of compression therapy applied to the selected body part/area; and generating compression garment fit information for the selected body part/area. In another example, a library of compression garment fit information includes anthropometric information generated for individuals in need of compression therapy; information associated with a health condition for the individuals; and compression garment fit instructions provided by a compression garment manufacturer.

A method for identifying, locating, and mapping targets of interest using unmanned aerial vehicle (UAV) camera footage in GPS-denied environments. In one embodiment, the method comprises obtaining UAV visual data, passing the UAV visual data through a convolutional neural network (CNN) in order to detect targets of interest based on visual features disposed in the UAV visual data, wherein the detection by the CNN defines reference points and pixel coordinates for the UAV visual data, applying a geometric transformation to known and defined pixel coordinates to obtain real-world orthogonal positions; and projecting the detected targets of interest onto an orthogonal map based on the obtained real-world orthogonal positions, all without GPS data.

The present disclosure provides an image processing method and apparatus, and relates to the field of image processing, and in particular to the field of image annotation. An implementation is: obtaining an image to be processed including a target region to be annotated; in response to a first click on the target region, performing a first operation to expand a predicted region for the target region based on a click position of the first click; in response to a second click in a position where the predicted region exceeds the target region, performing a second operation to reduce the predicted region based on a click position of the second click; and in response to determining that a difference between the predicted region and the target region meets a preset condition, obtaining an outline of the predicted region to annotate the target region.

Various examples are provided related to identification of anthropometric information for fitting of compression garments. In one example, a method of generating compression garment fit information includes acquiring images including a selected body part or body area in need of compression therapy; processing the acquired images along with a library of compression garment fit information; generating a 3D reconstruction of the selected body part/area; deriving anthropometric information for the selected body part/area from the 3D reconstruction; providing a compression value corresponding to a prescribed or intended amount of compression therapy applied to the selected body part/area; and generating compression garment fit information for the selected body part/area. In another example, a library of compression garment fit information includes anthropometric information generated for individuals in need of compression therapy; information associated with a health condition for the individuals; and compression garment fit instructions provided by a compression garment manufacturer.

A method for use of machine learning in computer-assisted anatomical prediction. The method includes identifying with a processor parameters in a plurality of training images to generate a training dataset, the training dataset having data linking the parameters to respective training images, training at least one machine learning algorithm based on the parameters in the training dataset and validating the trained machine learning algorithm, identifying with the processor digitized points on a plurality of anatomical landmarks in a radiographic image of a person's skeleton displayed on a screen by determining anatomical relationships of adjacent bony structures as well as dimensions of at least a portion of a body of the skeleton in the displayed image using the validated machine learning algorithm and a scale factor for the displayed image, and making an anatomical prediction of the person's skeletal alignment based on the determined anatomical dimensions and a known morphological relationship.

Automatic selection of region in 3D point cloud is provided. Neighbor points are determined for given seed point of seed points. Responsive to a color difference of a given neighbor point from given seed point being less than neighbor color distance threshold and responsive to an angle between a normal of given neighbor point and a normal of given seed point being less than neighbor normal angle threshold, given neighbor point is added to region in 3D point cloud. Responsive to curvature at given neighbor point being less than curvature threshold, responsive to color difference of given neighbor point from initial seed point being less than initial seed color distance threshold and responsive to an angle between a normal of given neighbor point and a normal of initial seed point being less than an initial seed normal angle, given neighbor point is added to seed points for processing.

A system and method utilizing an existing boresighted HUD and a camera to determine the transformational and rotational relationships between the camera image and the boresighted HUD image of an aircraft to properly orient the camera for landing operations. Known boresight features are identified in a HUD image and are related to the same boresight features in a camera image, and transformations are applied to align pixels in the camera image to the pixel locations of the boresight features in the HUD image. Object recognition algorithms may be employed to identify the boresight features in the camera image.

Techniques are described for contouring of a region of interest based on imaging parameters of spatial imaging data and guided by user input of locations in the spatial imaging data, which may be used for segmentation or radiation treatment planning. An approach is described of combining a new paint brush tool with an edge-detection algorithm to correct for both the jagged contours and the painting routine not being executed often enough. By using an edge-detection algorithm, the user does not need to focus as much attention on moving the mouse accurately because the system will find the true organ boundary (e.g., using the image gradient) automatically, which may also lead to more time savings.

A method of generating a user interface includes obtaining input data indicative of respective positions of a plurality of elements within an input plane and generating a model of a surface of a three-dimensional structure. The surface has circular symmetry and includes two curved polar caps and a curved equatorial belt, wherein a curvature of each of the polar caps is greater than a curvature of the equatorial belt. The method includes mapping the elements to respective positions on the model, and determining a position and orientation of a virtual camera, wherein the determined position of the virtual camera is exterior to the model. The method includes determining a field of view of the virtual camera containing a concave interior portion of the model, and rendering a projection of the model corresponding to the determined field of view and comprising elements mapped to positions on the concave interior portion.