50 results about "Nonlinear registration" patented technology

The invention relates to a multimodal medical image registration and visualization method. Based on FLTK, VTK, and IRTK technologies, the method provides a friendly operation interface and visualization functions, simplifies the complex algorithm parameter configuration process, facilitates registration operations, and realizes multi-modal medical image registration; provides image denoising preprocessing to Improve image quality and reduce the impact of noise on registration accuracy; in the case of large coordinate position differences between the target image and the source image, a point set registration method is provided to complete the coarse registration of the coordinate position, which provides a basis for subsequent rigid, affine and nonlinear registration. Accurate implementation for initialization; support cascade working mode of multiple registration methods to improve registration efficiency; use multi-threaded programming technology to realize real-time dynamic display of the registration process, which is convenient for understanding and monitoring the registration process; registration The result is automatically output and saved; using the transformation matrix of the registration result and adding a smooth limit item, the function of diffeomorphism transformation and fast segmentation can be realized.

The invention relates to a construction method for a health people white matter fiber tract atlas. The method comprises the steps that through seeking the corresponding relationship between individual gray matter structure space and DTI space, the information of the tensor image average space of group people is calculated, and the DTI data of individuals are transformed into the average space; parameters jointly registered by a magnetic resonance anatomical structure image and an MNI standard form are integrated with registration parameters in the group by using related integration means and applied in an individual T1 dataset, and a group transition tensor image template is established; the individual T1 data are registered into the transition template through non-linear registration, and transformation parameters of a T1 structure image registered to the transition template are obtained; the transformation parameters are used in the individual T1 data, the correction for direction of a tensor field on each voxel point is carried out by using methods of keeping the main characteristic direction, the linear average of the voxel point one by one is carried out on the corrected tensor, and finally the diffusion tensor atlas of the specific health people group is obtained.

The invention relates to the computation technology field for brain shrinkage degree, and more particularly to a quantitative calculation method for automatic regional brain shrinkage degree. The quantitative calculation method comprises at least the steps of brain template library creating, non-linear registration mapping and brain shrinkage degree calculation. The quantitative calculation method is convenient, rapid and highly quantitative. With the method, it is possible to find a highly feasible scheme for quantitative brain shrinkage degree so as to provide reference to other judgments based on the brain shrinkage degree.

The location of the tumor in the patient changes according to the breathing of the patient. Radiography of the region containing the tumor is performed at a number of different respiratory phases. The X-ray image deformation calculation unit (14) of the device calculates deformation vectors of other images with respect to the reference image by performing nonlinear registration between images in the reference respiratory phase and images in other respiratory phases. A position calculating unit (15) calculates data indicating the position of the tumor in each respiratory phase by applying a deformation vector to the data indicating the position of the tumor in the reference respiratory phase. On the other hand, a template matching unit (16) recognizes a region (83) matching a template image (81a-81n) including a tumor image (M) in an image (82) at each respiratory phase. The template matching unit (16) thus calculates data indicating the position of the tumor in each respiratory phase. The error of the output of the template matching unit (16) relative to the output of the position calculation unit (15) is obtained. Optimizing the parameters for template matching makes the error small.

The invention belongs to the field of medical image analysis and application, relates to a method for registering a stereoscopic target image to a template image, and specifically relates to the method for registering a stereoscopic medical image rapidly, accurately and non-linearly based on sparse representation. The sparse combination coefficient can be obtained by utilizing detected mark corresponding points of the template image and the target image and by searching an established deformation field dictionary and corresponding point dictionary. The coefficient can fuse corresponding examples of the deformation field dictionary, so that a final deformation field of the target image is obtained, and the target image is registered to the template image. The method has a relatively-good application in a clinic environment and can be used for registering brain nuclear magnetic resonance images rapidly and accurately in neurosciences, or used for accurately positioning the position of the prostate in prostate cancer radioactive therapy, and can realize more accurate registering more rapidly.

A cervical caner image automatic partition method based on T2 weighted magnetic resonance imaging (MRI) T2-MRI and dispersion weighted (DW)-MRI includes that a DW-MR image is registered to a T2-MR image by using a non-linear register method, and the registered DW-MR image is sorted; the T2-MR image is filtered through non-linear anisotropic diffusion filtering technology, a bladder and a rectum are partitioned, and an interested area is partitioned through a partition result of the bladder and the rectum; and a combined maximum a posterior (CMAP) method is adopted for an interested area of the T2-MR image and the DW-MR image to conduct precise partition of a tumor. The cervical caner image automatic partition method fully uses effective information of the T2-MR image and the DW-MR image, can effectively overcome effects of noise, partial volume effect and strength overlapping in the T2-MR image, is precise and effective, and has important clinical and application value on prevention, diagnosis and treatment of the cervical cancer.

The invention discloses an image analysis method, which comprises the steps of: obtaining a group of T1 weighted images including a control group image and an experimental group image, pre-processing all T1 weighted images, carrying out non-linear registration on preprocessed T1 weighted images into a preset template image, dividing the non-linear registered T1 weighted images to generate a plurality of interest regions, extracting at least one feature data of each interest region to obtain feature matrix information, calculating to obtain a reference value according to a preset Bayesian multi-layer model and the feature matrix information, and judging difference significance of the interest regions between the experimental group image and the control group image according to the relation between the reference value and a preset threshold value. The invention also provides a corresponding image analysis system. The image analysis method and the system can extract more and richer parameters to be analyzed and more diverse information in comparison with the conventional technology, and can simultaneously process multi-region multi-feature significance statistical test so that morphology analysis is more complete.

The present invention describes a method and the system for matching fingerprint images obtained from different fingerprint image capturing devices and implemented in an electronic device. Specifically, the method has steps of identifying minutiae and then identifying core point of the fingerprint images. Thereafter, a line of symmetry is determined and minutiae are identified by matching ridge count, location angle and orientation angle of a first image of the fingerprint images with a second image of the fingerprint images is identified. Thereafter, prominent minutiae from the candidate minutiae having three adjacent minutiae with same local structure are identified. Out of scope minutiae are eliminated by non-linear registration of fingerprint images based on ridge-count and position angle. Thereafter, matching score is generated based on number of candidate minutiae, prominent minutiae and common overlapped minutiae after non-linear registration of the second fingerprint image in comparison with the first fingerprint image for authentication.

The invention provides a space calibration method and system of a three-dimensional brain graph data set. The space calibration method of the three-dimensional brain graph data set comprises the following steps: S1, preliminary orientation correction; S2, feature segmentation; S3, down-sampling; S4, linear registration; S5, nonlinear registration; S6, up-sampling; and S7, linear and nonlinear correction. According to the method, problems of difficult space calibration and inaccurate calibration of three-dimensional brain graph data sets of different individuals, three-dimensional brain graph data sets of different modalities, three-dimensional brain graph data sets of different imaging quality and the like are overcome, a more objective, accurate and sufficient feature extraction manner isadopted, and space calibration of the three-dimensional brain graph data sets of the different individuals, the three-dimensional brain graph data sets of the different modalities, the three-dimensional brain graph data sets of the different imaging quality and the like is realized. The method has certain versatility.

The invention provides a spinneret plate special-shaped hole size detection method. The spinneret plate special-shaped hole size detection method comprises the following steps: based on a line segmentand arc combination mechanism during spinneret plate hole silk outlet surface outer contour machining design, adopting universal non-rigid parametric matching model based on line segment and arc combination to carry out modeling and detection on any spinneret plate special-shaped hole, achieving the universality of special-shaped hole detection. The method comprises the steps: carrying out imageextraction, contour analysis and feature extraction through a standard special-shaped hole image, and automatically generating a universal matching model; correcting the rotation angle of the image contour of the to-be-detected hole; carrying out global and local connected region pre-registration and parameterized model non-linear registration based on least squares, and optimizing the positions of the feature control points, optimally registering the model contour reconstructed by the non-rigid matching model to the contour position of the hole to be measured, and solving the hole quality monitoring parameters according to a preset hole size feature parameter calculation method.

The invention relates to a three-dimensional nonlinear registration method and system for a massive brain image data set. The three-dimensional nonlinear registration method of the massive brain imagedata set includes the following steps of: S0, down-sampling; S1, low-resolution image registration; and S2, high-resolution image fast transformation. The low-resolution image registration step further includes the following steps of S11, linear registration; and S12, nonlinear registration. The high-resolution image fast transformation step further includes the following steps of: S21, transformation matrix calculation for high-resolution registration; S22, post-high-resolution registration spatial range calculation; S23, blocking; and S24, high-resolution nonlinear transformation. Accordingto the three-dimensional nonlinear registration method and system for the massive brain image data set of the invention, the registration information of low-resolution image data is adopted to transform the massive high-resolution three-dimensional image data set, so that the nonlinear registration of the massive high-resolution three-dimensional image data set can be realized. The method and system of the invention are applicable to all linear registration based on transformation matrices and all nonlinear registration based on displacement fields and can quickly perform nonlinear registration on TB-level data sets.

The invention discloses a transcranial magnetic stimulation positioning system based on a deep effect brain region, and the system comprises the steps: obtaining the data of a tested magnetic resonance image, which comprises a resting state functional image and a high-resolution T1 structure image; performing data preprocessing on the function image and the structure image, and establishing an individual space; the deep effect brain region template is converted to an individual space through nonlinear registration, and a target region of interest is defined; calculating functional connections between voxels of the deep effect brain region and the cortical region by adopting a Granger causality analysis method based on wavelet transformation; and screening Top block masses according to the functional connection strength, calculating the gravity centers of the Top block masses, and determining individualized stimulation targets in combination with the skull surface distance. According to the method, spontaneous brain activity function variation of the brain of an individual and the brain signal propagation direction are fully considered, accurate positioning of different individual levels is achieved, relatively stable cortex stimulation targets are obtained, and therefore the clinical curative effect of TMS is optimized.

The invention discloses a standardized cerebrovascular feature extraction method and device and a medium, and relates to the field of cerebrovascular feature extraction, and the method comprises the steps: calculating the blood vessel length and the blood vessel volume based on a target cerebrovascular three-dimensional segmentation image, and determining a blood vessel center line based on Voronoi covariance measurement; according to the blood vessel center line, the blood vessel diameter, the blood vessel curvature radius, the blood vessel bending degree, the blood vessel bifurcation angle and the blood vessel bifurcation density are calculated; performing nonlinear registration on the target cerebrovascular three-dimensional segmented image and a preset standard brain template to obtain a plurality of target standard cerebrovascular partitions and individual feature results in each target standard cerebrovascular partition; and performing group level analysis on the target cerebrovascular three-dimensional segmented image based on a voxel morphological measurement method to obtain a group feature result. According to the invention, multi-level standardized, accurate and stable cerebrovascular feature extraction can be realized.

The invention belongs to the field of medical image analysis and application, relates to a method for registering a stereoscopic target image to a template image, and specifically relates to the method for registering a stereoscopic medical image rapidly, accurately and non-linearly based on sparse representation. The sparse combination coefficient can be obtained by utilizing detected mark corresponding points of the template image and the target image and by searching an established deformation field dictionary and corresponding point dictionary. The coefficient can fuse corresponding examples of the deformation field dictionary, so that a final deformation field of the target image is obtained, and the target image is registered to the template image. The method has a relatively-good application in a clinic environment and can be used for registering brain nuclear magnetic resonance images rapidly and accurately in neurosciences, or used for accurately positioning the position of the prostate in prostate cancer radioactive therapy, and can realize more accurate registering more rapidly.

The invention discloses a control device and computer equipment for a brain health therapeutic apparatus, and relates to the technical field of brain stimulation control, and the method comprises the steps: synchronously collecting an fMRI activity map, sMRI data and whole-brain voxel-level DTI data of a subject, extracting a cortex target region to be stimulated, and reconstructing and generating a triangular mesh model; non-linear registration is carried out on whole-brain voxel-level DTI data and the sMRI space, and a target region vertex direction field is generated; calculating a direction field average direction and a direction field covariance of the direction field at the vertex of the target region; if the covariance of the direction field does not exceed a preset covariance threshold value, controlling the brain health therapeutic apparatus to perform magnetic stimulation operation on the cortex target region according to the average direction of the direction field; and if the direction field covariance exceeds a covariance threshold, screening a first target region vertex sector and a second target region vertex sector, and controlling the brain health therapeutic apparatus to perform time division multiplexing vector scanning between sectors. Therefore, the accuracy of targeted stimulation and the individual suitability are greatly improved.

The non-linear registration method for remote sensing image adopts local triangular match lattice non-linear conversion method. Geographic image with geometrical distortion and non-rigid distortion image are registered, positioned and corrected. The registration and geometrical correction are performed in every triangular local image area with net point as vertex. For fast registration, adapting area are selected via man-power interaction and homonymic characteristic points are searched automatically. The registration between basic image and registered image re registered through the registration between several son images. Position registration and geometrical correction are realized with the son image lattice and image transformation inside the lattice.

The invention relates to an edge calculation enabling tunnel back break real-time detection method and system, and belongs to the technical field of tunnel construction intelligent monitoring and quality control, and the method comprises the following steps: 1, collecting data in real time through array multi-mode sensors deployed on a tunnel excavation surface; 2, dynamically constructing a local BIM model at an edge computing node, and carrying out noise reduction processing on the point cloud data by adopting an adaptive voxel filtering algorithm; step 3, carrying out nonlinear registration on the real-time point cloud and the design model based on an improved ICP (Inductively Coupled Plasma) algorithm; 4, extracting back-break features through a lightweight convolutional neural network, and dynamically updating network parameters by adopting a transfer learning framework; step 5, generating an augmented reality visualization report including the back break amount, the spatial distribution and the risk level on the edge side, and transmitting the augmented reality visualization report back to the control center through the low-delay communication module; the method has the beneficial effects that the real-time point cloud and the design model are subjected to nonlinear registration, and the requirement of tunneling deviation is monitored in real time.

A medical image data processing apparatus comprises a registration unit configured to perform a non-rigid registration of a first set of medical image data and a second set of medical image data, wherein the registration unit is further configured to determine, for at least one region, a rigid or affine registration that is an approximation of the non-rigid registration.

The invention discloses an acupoint registration method based on feature fusion of a midline and a hip edge, and relates to the field of graphic image processing, and the method comprises the steps: S1, constructing a standard coordinate system matched with the acupoints of a human body; s2, acquiring a depth image based on a structured light camera to generate three-dimensional point cloud data of the back of the human body; s3, recognizing a human body median line from the three-dimensional point cloud data based on symmetric analysis; s4, identifying a hip edge from the three-dimensional point cloud data based on the detection gradient extreme value; s5, establishing a local human body coordinate system based on the human body median line and the hip edge; and S6, registering the local human body coordinate system with the standard coordinate system by using a non-linear registration algorithm, and mapping the registered coordinates into the world coordinate system of the user to finish the acupuncture point registration of the crowds with different body types. The invention provides an acupoint registration method based on feature fusion of a median line and a hip edge, which is used for reducing errors caused by body type differences and improving the accuracy of acupoint recognition at the same time.

The invention relates to the technical field of brain disease target positioning, in particular to a brain disease target positioning system and method based on functional magnetic resonance imaging. The method comprises the following steps: calculating a whole brain function connection map of an individual patient based on 4D rs-fMRI data, calculating a Z score of each voxel function connection strength in the whole brain function connection map of the individual patient, and screening abnormal voxels based on a preset threshold value; aggregating the abnormal voxels through a clustering algorithm so as to determine a cluster; the therapeutic effect scores of the multiple clusters can be predicted through the machine learning model, the clusters and the candidate therapeutic targets can be screened based on the therapeutic effect scores, the positions of the candidate targets can be accurately determined, and multiple selections of the candidate targets can be provided; the target spot is converted into an accurate target spot suitable for an individual through a nonlinear registration algorithm; and determining a target positioning state according to the deviation between the accurate target and the actual treatment target, and adjusting corresponding parameters based on the target positioning state. The invention provides a plurality of targets and effectively determines the positions of the targets.

The application carries out quantitative susceptibility mapping (QSM) reconstruction on original gradient echo data (GRE), then constructs an individual space standardization QSM brain template by using structural images, carries out linear and nonlinear registration processing on QSM images of all subjects and the brain template, so as to achieve the purpose of standardization processing of the QSM images; meanwhile, the structural images are automatically segmented, the gray matter and white matter masks obtained after segmentation are used to carry out matrix point multiplication on the QSM images, then the standardized QSM images are smoothed and corresponding smoothing compensation is carried out; in FSL, a general linear model is constructed, and voxel-level whole brain statistical analysis is carried out on the QSM images; the masks obtained after automatic nucleus segmentation are used to extract the susceptibility values of corresponding deep nuclei in the QSM images; the whole process does not need to manually draw the interested region, and the quantitative analysis of the susceptibility values based on voxels and based on the interested region can be automatically completed, and the repeatability, comparability and reliability are greatly improved.

Disclosed in the present invention is a nonlinear registration optimization method for a fastener point cloud. The method comprises the following steps: S1, collecting original data; S2, preprocessing the data; S3, converting frame images into a deformation map; S4, performing registration between the frame images; and S5, simplifying corrected point cloud data. The method in the present invention improves the data accuracy and confidence level of an original point cloud, and provides more reliable data input for subsequent point cloud processing and advanced understanding processes, thereby ensuring the achievement rate of point cloud processing objectives; and the point cloud optimization process in the present invention has high robustness, and effective features of original point cloud data are retained to the greatest extent while noise points and outlier points are eliminated, thereby achieving high detail retainability. Moreover, the nonlinear optimization process avoids falling into the trap of local optimization, achieving high computational efficiency, strong universality, and high potential for popularization and application.

The present application discloses a method, device and medium for extracting standardized cerebrovascular features, relating to the field of cerebrovascular feature extraction. The method includes: calculating the blood vessel length and blood vessel volume based on the three-dimensional segmentation image of the target cerebrovascular vessels, and determining the blood vessel centerline based on the Voronoi covariance metric; calculating the blood vessel diameter, blood vessel curvature radius, blood vessel bending degree, blood vessel bifurcation angle and blood vessel bifurcation density respectively according to the blood vessel centerline; performing non-linear registration on the three-dimensional segmentation image of the target cerebrovascular vessels and a preset standard brain template to obtain a plurality of target standard cerebrovascular partitions and the individual feature results within each target standard cerebrovascular partition; performing population-level analysis on the three-dimensional segmentation image of the target cerebrovascular vessels by means of voxel-based morphological measurement to obtain population feature results. The present application can achieve multi-level standardized, accurate and stable cerebrovascular feature extraction.

The invention discloses a multi-modal image registration method and device based on a generative model and self-supervised learning, and a medium, and the method comprises the steps: carrying out the modal conversion of a source image through employing a Pix2Pix-Transform generative model, extracting global and local features from the source image and a target image through a DINO model in an image registration process, carrying out the region division of the image based on the features, and carrying out the recognition of a target image through a DINO model; for each region, weighted distribution is carried out according to the feature intensity and importance of the region, so that the key region can obtain higher weight and precision in the registration process, and the weight of the background region can be properly reduced. A region self-adaptive registration technology is adopted, high-precision non-linear registration is carried out on a key region, low-precision linear registration is carried out on a background region, and the overall registration efficiency is ensured. According to the self-adaptive weighted registration method, the overall precision and efficiency of image registration are effectively improved, and particularly, when a multi-modal image is processed, the registration difficulty caused by modal difference can be remarkably reduced.

The application discloses a kind of slice image registration and marking method, belong to image processing technical field, steps are as follows: based on generative adversarial network, using the first generator of well-trained brain slice image is carried out style migration, obtain the brain slice image after style migration;Improved ConvNext neural network model is used to the brain slice image after style migration carries out atlas level position identification, and is matched with the allen sign picture slice image, constitutes brain slice image matching pair;Several brain slice image matching pairs are input based on Voxelmorph registration model, obtain the nonlinear registration vector field between brain slice image and allen sign picture slice image;Based on nonlinear registration vector field, the atlas information of allen standard atlas slice image is registered to brain slice image, complete brain slice image registration and solve academic annotation.The application solves the problem that mouse brain slice image registration and marking accuracy and efficiency are insufficient.

The present invention provides a map-assisted drawing method, a computer-readable storage medium and an electronic device, which includes steps S1, constructing an axial correspondence relationship of a brain slice; step S2, normalizing old map data; step S3, extracting regional features; step S4, nonlinear registration; and step S5, integrating old map information, so as to achieve accurate matching of brain region positions and boundary information on the old map to the image of a new map to be drawn.

An automatic segmentation method for a cervical cancer image based on T2-weighted magnetic resonance imaging (T2-MRI) and diffusion-weighted magnetic resonance imaging (DW-MRI), including: registering a DW-MR image to a T2-MR image using a nonlinear registration method, and classifying the registered DW-MR image; filtering the T2-MR image using a nonlinear anisotropic diffusion filtering technique, segmenting a bladder and a rectum, and segmenting a region of interest using the segmentation results of the bladder and the rectum; and performing accurate segmentation of a tumour on the region of interest of the T2-MR image and the DW-MR image using a combined maximum a posteriori probability (CMAP) method. The present invention makes full use of the effective information about a T2-MR image and a DW-MR image, can effectively overcome the influences of noise, partial volume effect and intensity overlap in the T2-MR image, is an accurate and effective segmentation method for cervical cancer, and has important clinical significance and application value for the prevention, diagnosis and treatment of cervical cancer.

The invention belongs to the technical field of flaw detection, and discloses an intelligent detection method and system for surface flaws of a printed circuit board based on a dynamic template, and the method comprises the steps: obtaining a design file and a surface image of a to-be-detected printed circuit board; performing hierarchical analysis on the data to obtain multiple pieces of hierarchical information, and performing alignment processing on the multiple pieces of hierarchical information to obtain standardized hierarchical data; performing feature enhancement based on the standardized hierarchical data, fusing the enhanced hierarchical data, and constructing a dynamic detection template based on the fused data; carrying out image preprocessing on the collected surface image, and carrying out nonlinear registration on the surface image after image preprocessing in combination with a dynamic detection template to obtain a corresponding registration image; processing the obtained registration image to obtain a flaw detection and classification result; according to the method, the local deformation model and the feature point incidence matrix are accurately established, so that the accuracy of nonlinear registration is ensured, and the problem of false detection caused by deformation in printed circuit board detection is effectively solved.

Disclosed in the present invention is a nonlinear registration optimization method for a fastener point cloud, comprising the following steps: S1, collecting raw data; S2, preprocessing the data; S3, transforming frame images into a deformation graph; S4, performing registration between the frame images; and S5, simplifying corrected point cloud data. The method of the present invention improves the accuracy and confidence score of raw point cloud data, and provides a more reliable data input for subsequent point cloud processing and higher-level interpretation processes, thereby ensuring the fulfillment rate of point cloud processing objectives. The point cloud optimization process of the present invention exhibits high robustness, preserves effective features of the raw point cloud data as much as possible while removing noise points and outliers, and highly preserves fine details. Additionally, the nonlinear optimization process avoids getting trapped in local minima, exhibits high computing efficiency, demonstrates high universal applicability, and has broad potential for practical application.