49 results about "Brain White Matter" patented technology

Provided is a white matter fiber brain map construction method by means of a diffusion tensor imaging medical image. The method comprises the steps that an original diffusion tensor imaging image is input (101); image preprocessing is conducted on a diffusion tensor image (106), wherein the substeps of data format conversion (102), 4D image converting (103), anisotropy value computing (104) and bone removing operation (105) are included; a registration technology is utilized (107), reflection transformation (108) and a large-deformation differential homomorphic mapping algorithm (109) are used, and different tested diffusion tensor images are registrated to the same standardized space (107); diffusion weighted imaging analysis is conducted (113), wherein the substeps of analyzing a deformation field by means of a diffusion tensor model (110), conducting diffusion weighted imaging redirection (111) and estimating diffusion tensor (112) are included; two brain white matter maps are constructed for patients and normal people (115); machine learning is conducted (114), wherein feature extraction (116) and feature selection (117) are conducted on tested pictures, data sets are divided into a training set and a test set, machine learning is conducted according to the maps (118), and training is conducted to generate a classifier (121).

The invention relates to a mixed brain white matter nerve fiber automatic cluster and a marking method. The marking method is characterized in that: firstly, using a high dimensional elastic registration method to match a standard brain atlas to a single brain space, thereby being capable of dividing a single brain into 104 anatomical regions with structural information; secondly, using differentnerve fiber bundles to extract and mark 15 nerve fiber bundles according to the characteristics of different brain anatomical regions; thirdly, to other 4 nerve fiber bundles which can not be extracted by the above method due to strong winding between the 4 nerve fiber bundles and the nerve fiber bundles at the periphery thereof, using a method combing Kernel-PCA and Fuzzyc-mean and based on similarity measurement to automatically carry out clustering analysis; and at last, using an identification method based on characteristics to mark the result of clustering analysis. By the above method, 19 main white matter nerve fiber bundles of the brain can be extracted and marked automatically.

In the invention, through an improved threshold segmentation algorithm, segmentation of a grey matter and a white matter of a brain image is realized, which is good for auxiliary diagnosis of brain atrophy of a patient. The method is characterized by firstly, using an adaptive filter to process an original image, reducing a noise of the original image and enhancing a contrast ratio of the image; then, for the processed image, using a region growing algorithm to carry out brain decollement on the image, rejecting a non-parenchymal portion; and then using an iterative threshold method to calculate a threshold and using a near weighted value algorithm to carry out binarization segmentation on the image so that an obtained result is a segmentation result of a brain white matter; finally, based on a result of the previous step, carrying out segmentation so that the obtained result is a segmentation result of a brain grey matter.

The invention discloses a brain white matter high signal detection and positioning method based on multiple maps. The method comprises the following steps of firstly, establishing a multi-map libraryby using FLAIR data of brains of normal old people; secondly, registering the maps in the map library to a target image one by one, and using a multi-map fusion algorithm for fusion; and finally, realizing image segmentation according to a fusion result, and automatically detecting voxels with local abnormal intensity as white-quality high signals. According to the method, the while the brain white matter is segmented, the white matter high-signal area is also detected and positioned, the area is well matched with a manual description result, the higher accuracy and the precision are achieved,and the effect is superior to that of other white matter high-signal detection methods at present.

The invention discloses a white matter high signal segmentation method, device and apparatus and a storage medium. The method comprises the steps of obtaining a to-be-segmented image; performing whitematter high signal segmentation on the to-be-segmented image based on a first segmentation network to obtain a white matter high signal label prediction result of the to-be-segmented image; and taking the to-be-segmented image and the white matter high signal label prediction result of the to-be-segmented image as input of a second segmentation network, and correcting the white matter high signallabel prediction result of the to-be-segmented image based on the second segmentation network to obtain a white matter high signal label correction result of the to-be-segmented image. By utilizing the technical scheme provided by the invention, the white matter high signal segmentation result with higher resolution, more accurate boundary and better spatial continuity can be obtained, and the problem of fuzzy boundary when the first segmentation network segments the white matter high signal is solved.

The invention discloses a genetic image multi-modal fusion analysis method based on white matter integrity and DNA methylation. The method comprises the following steps: analyzing the whole-brain white matter fiber integrity FA value of a subject, performing inter-group comparison on a patient suffering from a certain neuropsychiatric disease and a normal person, typing a difference region name in an Allen brain library to obtain names of genes highly expressed in the brain regions, extracting methylation levels of CpG sites corresponding to the genes from a genome methylation sample, and counting the correlation between the methylation level of the CpG sites and the FA value by adopting correlation analysis. According to the method, multi-modal fusion analysis is carried out on the brain structure development mode of the neuropsychiatric diseases and the epigenetic basis behind the neuropsychiatric diseases, the analysis method ingeniously combines FA and DNA methylation two-modal data for analysis, and great application and guidance values are brought to deep mining of existing image and epigenetic multi-modal data information.

The invention discloses a white matter high-signal segmentation method based on multi-scale fusion and attention splitting, and belongs to the technical field of brain magnetic resonance image processing. According to the technical key points, the method comprises the following steps: obtaining a white matter high-signal FLAIR image data set, and carrying out the division and preprocessing of the data set; constructing and training the white matter high-signal segmentation model based on multi-scale fusion and split attention, and obtaining the white matter high-signal segmentation model when the training meets a termination condition; and inputting each image in the test set into the trained white matter high-signal segmentation model based on multi-scale fusion and split attention for testing. The beneficial effects are that the white matter high-signal segmentation method based on multi-scale fusion and split attention can effectively improve the WMH segmentation accuracy, and especially has good recognition capability for small lesions. The method has reference significance for segmentation of other medical images and also has positive significance for promoting computer-aided diagnosis of brain diseases.

The invention discloses a magnetic resonance diffusion tensor brain image analysis method and system for newborns, and belongs to the field of medical image processing. The method comprises the following steps: A, preprocessing a newborn diffusion tensor imaging DTI image; b, dispersion tensor parameter calculation: fitting the intrinsic value of the dispersion tensor through a DTI tensor model, and calculating a dispersion parameter graph based on an anisotropy fraction FA, an average diffusion coefficient MD, an axial diffusion coefficient AD and a radial diffusion coefficient RD; c, performing image registration; and D, extracting a skeleton based on the anisotropy fraction FA, and projecting the skeleton based on the anisotropy fraction FA, the average diffusion coefficient MD, the axial diffusion coefficient AD and the radial diffusion coefficient RD of the testee onto the skeleton. According to the method, the newborn DTI image is subjected to preprocessing, diffusion tensor parameter calculation, image registration and FA skeleton extraction, and finally the brain image with clear brain white matter and brain gray matter signals is obtained.

The invention discloses an automatic quantitative analysis system and an interpretation method for white matter lesions, which utilize an artificial intelligence method to automatically and accurately sketch lesions, accurately calculate the volume of the white matter lesions and provide possibility for accurate evaluation of disease progression. Meanwhile, the invention provides the automatic interpretation system, multi-dimensional information such as anatomical positioning, morphological structure, lesion size, lesion occurrence time and bleeding of lesions is determined through human-computer interaction, lesion positioning visualization, image index quantification, report term standardization and operation interface friendliness are achieved, logic analysis is automatically made, disease judgment is accurately obtained, and inconformity between a conclusion and description and missing report of important conclusions are avoided.

The invention discloses a newborn brain development quantitative analysis method based on a magnetic resonance diffusion tensor brain image. The method comprises the following steps: A, preprocessinga newborn DTI image, which includes data format conversion, image correction and brain tissue extraction; B, carrying out dispersion tensor parameter calculation; c, registering the FA image to an MNIstandard space and calculating an average FA image of the FA image; d, performing symmetry processing on the average FA image; e, extracting an FA skeleton from the average FA image subjected to thesymmetric processing; f, projecting each tested FA image into a symmetrical FA skeleton to obtain each tested FA skeleton diagram; and G, subtracting the FA values of the homologous isobaric voxels ofthe left brain and the right brain to obtain an absolute value of a difference value. By the adoption of the method, quantitative calculation is conducted on the brain white matter asymmetry index based on the DTI image, so that quantitative analysis on the brain development condition of the newborn is achieved.

The invention discloses a method for optimizing infant brain T1 weighted magnetic resonance imaging. The method includes, firstly, collecting T1 and PD mapping of the brain of infants aged 0-12 monthsto obtain the average T1 and PD values of the infants' white matter and infants' gray matter, and dividing the infants into three-month-old groups according to the characteristics of the relationshipbetween the T1 value of infants' white matter and the T1 value of gray matter; then, calculating the theoretical signal intensities of infants' white matter and gray matter in 3D T1 weighted images through Bloch simulation, and respectively determining an optimal theoretical TI optimization scheme for each group according to the contrast characteristics of infants' white matter/gray matter underdifferent TI; and finally, applying the optimal theoretical TI optimization scheme to the target infant's brain for 3D T1 weighted magnetic resonance imaging. According to the method, the blank of brain T1 weighted imaging optimization in the whole infant period from 0 to 12 months after birth is filled, the infants are divided into three-month-old groups according to the characteristics of the relationship between the T1 values of infants' white matter and the T1 value of infants' gray matter, and the optimal TI optimization schemes of different-month-old groups are respectively found, so that the T1 weighted imaging contrast of the infants' brain is obviously improved.

The invention is applicable to the technical field of image processing, and provides a brain white matter region segmentation method and device for a craniocerebral ultrasonic image and electronic equipment. The method comprises the following steps: carrying out diffusion and enhancement on an original input ultrasonic image, and then carrying out rough segmentation on the processed ultrasonic image to obtain a target image containing a brain white matter area and other non-interested areas; and carrying out secondary segmentation on a narrowed area, and removing the non-interested areas in the frame to obtain an accurately segmented brain white matter area. The invention can effectively avoid the problems that there is no clear boundary between an interested area and the surrounding environment in an ultrasonic image, the proportion of a brain white matter area is too low, and veins and other highlight areas have very strong negative effects on a segmentation result. Meanwhile, the white matter area segmentation device of the craniocerebral ultrasonic image and the electronic equipment provided by the invention can also achieve the same technical effect.

The invention discloses a human brain magnetic susceptibility tensor imaging method based on a cross mode, which can obtain phase data in multiple directions by rotating head postures when magnetic resonance imaging of a GRE sequence is carried out, so that the magnetic susceptibility and anisotropy information, namely magnetic susceptibility tensor imaging, of tissues can be deduced. In clinicalapplication, the scanning time and the imaging space limit the direction number and the direction angle of human brain imaging, so that the anti-interference capability of magnetic susceptibility tensor imaging is poor, which is an inverse problem of serious ill-conditioned conditions. According to the invention, a tensor spectrum decomposition technology is utilized; high-resolution magnetic susceptibility imaging and low-resolution diffusion imaging are combined; the invention provides a magnetic susceptibility tensor imaging method based on a cross mode. Phase data collected in six different head directions are used. Accurate calculation of the magnetic susceptibility tensor and the derivation amount is achieved, the method can be used for brain tissue microstructure characteristic evaluation, the derived fiber directional diagram is beneficial for exploring a novel living body nondestructive testing method of internal fiber characteristics of brain white matter and gray matter, andthe method has important significance in brain scientific research and brain disease diagnosis.

The invention discloses a white matter high signal extraction method and system based on a lesion diffusion algorithm. The method comprises the following steps: inputting an original image comprising an FLAIR image, a T1 weighted image and a white matter tissue prior probability graph; carrying out preprocessing operation on the input original image; performing abnormal signal product operation on the preprocessed image to obtain a tissue signal intensity distribution confidence map, and performing initialization and white matter high signal segmentation on the obtained tissue signal intensity distribution confidence map to obtain a white matter high signal probability map; and carrying out image binaryzation on the white matter high-signal probability graph generated after segmentation processing through a threshold value to obtain a white matter high-signal extraction graph, and superposing the white matter high-signal extraction graph on the standardized original image to obtain a white matter high-signal positioning graph. The white matter high signal extraction method is easy to operate, high in generalization, stable in effect, high in practicability and very effective.

The invention provides a magnetic resonance diffusion tensor imaging quality control comprehensive test phantom. The magnetic resonance diffusion tensor imaging quality control comprehensive test phantom comprises a brain white matter fiber bundle simulation area located on the upper portion and an imaging quality evaluation test area located on the lower portion. The brain white matter fiber bundle simulation area comprises a first shell, a test layer is arranged in the first shell, the test layer is filled with gel, and the gel is used for wrapping a fiber material to construct different binding densities and directions so as to form different hydrogen proton dispersion gradient coefficients to simulate the actual brain white matter fiber bundle trend and distribution of the human brain;the imaging quality evaluation test area comprises a second shell, and cylinders filled with the gel with set density are arranged in the second shell and used for simulating magnetic resonance to image soft tissues; four cylinders with set diameters are arranged in the second shell, each cylinder is filled with the gel with set density and granulated sugar, and a magnetic resonance diffusion tensor imaging model used for evaluating the signal-to-noise ratio and the apparent diffusion coefficient is constructed.

A depression classification method based on self-supervised learning and transfer learning is characterized by tracking whole-brain white matter fiber bundles based on the diffusion tensor imaging data, constructing a whole-brain white matter fiber bundle network, and automatically learning the brain network node representation and the network representation irrelevant to downstream tasks through the comparative learning by using a self-supervised learning strategy; constructing a group network based on the brain network characterization and the non-image phenotype information, converting a depression classification problem into a network node classification problem, and classifying the depression patients and normal contrast by using a graph convolution neural network model based on spectrogram convolution. According to the method, the self-supervised learning and the transfer learning are utilized, the problem that depression samples are fewer is partially solved, the brain network level features related to depression are effectively mined, and the depression classification precision is improved.

The invention discloses a brain white matter fiber tracking method based on a bidirectional ant colony algorithm of a communication mechanism, and the method employs the bidirectional ant colony algorithm of the communication mechanism, and employs two ant colonies for the independent path tracking of nerve fibers. On one hand, the ant movement selection process is divided into linear extension and movement according to pheromones according to roulette, the introduction of the linear extension enhances the time effectiveness of the algorithm, and the path optimization capability is improved according to the movement of the pheromones. On the other hand, when algorithm iteration falls into local optimum, pheromone volatilization coefficients of the two ant colonies are exchanged, the original search state of the ant colonies is changed, and the limitation of local optimum is broken through. The fiber tracking time is greatly shortened, and the nerve fiber searching space and searching efficiency are improved. In addition, the invention further provides a brain white matter fiber tracking device and equipment based on the bidirectional ant colony algorithm of the communication mechanism and a readable storage medium, and the technical effect of the brain white matter fiber tracking device corresponds to the technical effect of the method.

The invention provides a method of segmenting a brain white matter high signal based on a deep learning method. The method comprises the following steps of: constructing an original input image, carrying out preprocessing operation on the original input image, the preprocessing operation comprising image registration, first extraction of a brain contour, second extraction of the brain contour, image data standardization and image size unification, carrying out white matter high signal segmentation processing on the original input image after the preprocessing operation, and performing binarization processing on the brain white matter high signal probability graph generated after segmentation processing through a preset threshold to obtain a brain white matter high signal graph, and restoring the brain white matter high signal graph to an original size. The method has no limitation on the size of an input image, and is extremely high in segmentation speed, high in generalization and easy to use. The invention further provides a device for segmenting the brain white matter high signal based on the deep learning method.

The embodiment of the invention discloses a white matter high-signal segmentation method, device and equipment and a storage medium, and the method comprises the steps: obtaining target brain nuclear magnetic resonance image data, and carrying out the standardization processing of the nuclear magnetic resonance image data, so as to obtain target data; inputting the target data into a preset white matter high-signal segmentation model for segmentation processing to obtain white matter high-signal labeled data; and determining the position and the volume of the white matter high signal in the brain according to the white matter high signal mark data. By adopting the means, the nuclear magnetic resonance image data is subjected to standardization processing, so that the segmentation processing efficiency of the model can be improved, and automatic and reliable segmentation of the WMH in different MRI equipment images can be realized.