85 results about "3d registration" patented technology

An image-guided radiosurgery method and system are presented that use 2D/3D image registration to keep the radiosurgical beams properly focused onto a treatment target. A pre-treatment 3D scan of the target is generated at or near treatment planning time. A set of 2D DRRs are generated, based on the pre-treatment 3D scan. At least one 2D x-ray image of the target is generated in near real time during treatment. The DRRs are registered with the x-ray images, by computing a set of 3D transformation parameters that represent the change in target position between the 3D scan and the x-ray images. The relative position of the radiosurgical beams and the target is continuously adjusted in near real time in accordance with the 3D transformation parameters. A hierarchical and iterative 2D/3D registration algorithm is used, in which the transformation parameters that are in-plane with respect to the image plane of the x-ray images are computed separately from the out-of-plane transformation parameters.

A method and apparatus for performing 2D to 3D registration includes an initialization step and a refinement step. The initialization step is directed to identifying an orientation and a position by knowing orientation information where data images are captured and by identifying centers of relevant bodies. The refinement step uses normalized mutual information and pattern intensity algorithms to register the 2D image to the 3D volume.

The invention relates to a method and a device for registering 2D projection images of an object relative to a 3D image data record of the same object, in which, from just a few 2D projection images, a 3D feature contained in an object, which is also identifiable in the 3D images, is symbolically reconstructed. The 3D feature obtained in this way is then registered by 3D-3D registration with the 3D image data record.

A system architecture is disclosed that facilitates rapid execution of 3D registration or alignment algorithms. A first image module (e.g., RAM) is included to store data corresponding to images to be registered. A processor coupled to the first storage module accesses the data and determined mutual histogram (MH) values which are then used to compute mutual information (MI) between the images. The processor accumulates the MH values in a second image module. The second storage module is accessible so that registered images can be displayed. The architecture is scalable facilitating distributed calculations to speed-up the registration process.

A method for non-rigid registration of digital 3D coronary artery models with 2D fluoroscopic images during a cardiac intervention includes providing a digitized 3D centerline representation of a coronary artery tree that comprises a set of S segments composed of Q_S 3D control points, globally aligning the 3D centerline to at least two 2D fluoroscopic images, and non-rigidly registering the 3D centerline to the at least two 2D fluoroscopic images by minimizing an energy functional that includes a summation of square differences between reconstructed centerline points and registered centerline points, a summation of squared 3D registration vectors, a summation of squared derivative 3D registration vectors, and a myocardial branch energy. The non-rigid registration of the 3D centerline is represented as a set of 3D translation vectors r_s,q that are applied to corresponding centerline points x_s,q in a coordinate system of the 3D centerline.

A method and apparatus for convolutional neural network (CNN) regression based 2D/3D registration of medical images is disclosed. A parameter space zone is determined based on transformation parameters corresponding to a digitally reconstructed radiograph (DRR) generated from the 3D medical image. Local image residual (LIR) features are calculated from local patches of the DRR and the X-ray image based on a set of 3D points in the 3D medical image extracted for the determined parameter space zone. Updated transformation parameters are calculated based on the LIR features using a hierarchical series of regressors trained for the determined parameter space zone. The hierarchical series of regressors includes a plurality of regressors each of which calculates updates for a respective subset of the transformation parameters.

The present invention relates to a method of rapidly and repeatably bringing sharp objects into close proximity to a particular region of interest of a sample with high precision and accuracy in two or three dimensions using a laser guided tip approach with three dimensional registration to the surface.

A vision system of a vehicle includes at least one camera configured to be disposed at a vehicle so as to have a field of view exterior of the vehicle. Responsive to image processing of captured image data and with the at least one camera disposed at the vehicle, the image processor determines a three dimensional object present in the field of view of the camera and determines a point of interest on the determined object. The vision system uses triangulation to determine an estimated location in three dimensional space of the determined point of interest. The vision system processes additional frames of captured image data to enhance the estimation of the location in three dimensional space of the determined point of interest. The image processor is operable to estimate a distance to the determined object by comparing multiple frames of captured image data.

The present invention relates to minimally invasive X-ray guided interventions, in particular to an image processing and rendering system and a method for improving visibility and supporting automatic detection and tracking of interventional tools that are used in electrophysiological procedures. According to the invention, this is accomplished by calculating differences between 2D projected image data of a preoperatively acquired 3D voxel volume showing a specific anatomical region of interest or a pathological abnormality (e.g. an intracranial arterial stenosis, an aneurysm of a cerebral, pulmonary or coronary artery branch, a gastric carcinoma or sarcoma, etc.) in a tissue of a patient's body and intraoperatively recorded 2D fluoroscopic images showing the aforementioned objects in the interior of said patient's body, wherein said 3D voxel volume has been generated in the scope of a computed tomography, magnet resonance imaging or 3D rotational angiography based image acquisition procedure and said 2D fluoroscopic images have been co-registered with the 2D projected image data. After registration of the projected 3D data with each of said X-ray images, comparison of the 2D projected image data with the 2D fluoroscopic images—based on the resulting difference images—allows removing common patterns and thus enhancing the visibility of interventional instruments which are inserted into a pathological tissue region, a blood vessel segment or any other region of interest in the interior of the patient's body. Automatic image processing methods to detect and track those instruments are also made easier and more robust by this invention. Once the 2D-3D registration is completed for a given view, all the changes in the system geometry of an X-ray system used for generating said fluoroscopic images can be applied to a registration matrix. Hence, use of said method as claimed is not limited to the same X-ray view during the whole procedure.

Methods for updating a 2D/3D registration between a three-dimensional image data set corresponding to a target area subjected to a movement and a plurality of two-dimensional projection images of the target area include: selecting a plurality of contour points along a contour, the contour being visible in a first projection image and a three-dimensional image data set registered therewith, the plurality of contour points being associated with a rigid object in the target area; determining a displacement of each contour point of the plurality of contour points between the first projection image and a successively captured second projection image; obtaining movement information for each contour point of the plurality of contour points based on the determined displacement, the movement information describing a movement of the rigid object; and updating the registration based on the obtained movement information.

The invention provides an angle rotation real-time matching method based on try wearing of 3D (three dimensional) glasses. The angle rotation real-time matching method based on the try wearing of the 3D glasses includes steps: S1, capturing a picture of a human face through a camera, performing 3D registration on the human face, and building an initial 3D coordinate of the human face; S2, obtaining an initial glasses model through modeling; S3, scaling, shifting and rotating the initial 3D coordinate of the human face so as to obtain a standard human face image; S4, scaling, shifting and rotating the initial glasses model along with changes of the initial 3D coordinate of the human face in real time so as to obtain a standard glasses model; S5, placing the standard glasses model on the standard human face image so as to achieve image synthesis; S6, superposing the standard glasses model and the standard human face image so as to generate a demonstration image. The angle rotation real-time matching method based on the try wearing of the 3D glasses can achieve real-time following and matching of the try wearing of the 3D glasses even when the human face is rotated at a small angle, does not need a special device, and is convenient to use, low in cost, and good in effect.

A method of determining bone fragment navigation may include receiving pre-operative 2D image data of a reference bone structure and a bone fragment. The reference bone structure may include a first set of fiducial markers provided thereon, and the bone fragment may include a second set of fiducial markers provided thereon. The method may further include performing a 2D-3D registration between the pre-operative 2D image data and a 3D model of the reference bone structure and the bone fragment, after manual repositioning of the bone fragment, receiving second 2D image data, performing 2D-2D registration of the first set of fiducial markers and the second set of fiducial markers between the pre-operative 2D image data and the second 2D image data, and determining 3D movement of the bone fragment based at least in part on the 2D-2D registration.

A method for deformable registration including determining a vector field from a two-dimensional matching of a volume of an object of interest and a two-dimensional image of the object of interest, providing a deformation profile, and finding a volume deformation that maps to a state of the two-dimensional image, wherein the deformation is parameterized by the vector field and control points of the deformation profile to find a control point configuration of the volume deformation.

Registration of preoperatively acquired MRI images of soft parts to intraoperatively acquired X-ray images of soft parts is not possible. The invention shows a way of nevertheless using such preoperatively acquired images for superimposition with 2D projections of the soft parts, taking an indirect route via 3D/3D registration of images of the spinal column. For this purpose, 3D image data sets of the spinal column must be obtained separately on the one hand using MRI and on the other using the X-ray imaging system so that the 3D/3D registration produces a mapping rule which then also applies to the preoperatively acquired images of the soft part if the soft part images and the spinal images are acquired without intervening change in the patient position in the MRI scanner.

Systems and methods for assisting a physician in a medical intervention comprises performing a 2D-3D deformable registration, and more particularly, performing a 2D-3D registration based on multiple live 2D fluoroscopic views, and implemented on a multi-core processing framework such as a Graphics Processing Unit.

A method includes, following specification of an initial transformation as a test transformation that is to be optimized, determining a 2D gradient x-ray image and a 3D gradient dataset of the image dataset, carrying out, for each image element of the gradient comparison image, a check for selection as a contour point, and determining an environment best corresponding to a local environment of the contour point and extending around a comparison point in the gradient x-ray image for all contour points in the at least one gradient comparison image. Local 2D displacement information is determined by comparing the contour points with the associated comparison points, and motion parameters of a 3D motion model describing a movement of the target region between the acquisition of the image dataset and the x-ray image are determined from the displacement information and a registration transformation describing the registration.

The invention discloses an image real-time registration method and system. The method comprises the following steps of (1) acquiring data of a to-be-registered 2D image; and (2) inputting the data to a registration neural network obtained according to training of a target 3D image, and obtaining position information of the 2D image in the target 3D image, wherein the position information comprises an eigenvector of the to-be-registered 2D image in a coordinate system of the target 3D image and a distance between the to-be-registered 2D image and an origin of the coordinate system, and the eigenvector is a vector passing through the origin of the coordinate system and perpendicular to a 2D image plane, and comprises two parameters including an elevation angle and an azimuth angle. The system comprises an image acquisition module and a registration module. According to the method and the system, deep learning is introduced in the 2D-3D registration problem, a corresponding relationship between 2D and 3D is expressed as a deep convolutional neural network, and a policy for solving the exhaustion calculation problem is explored theoretically, so that the purpose of real-time registration is achieved.