19 results about "Hyperintensity" patented technology

A low field magnetic resonance imaging (MRI) stroke identification sequence uses an inversion recovery sequence. TE is set to the minimum or near-minimum value achievable by a system to minimize an impact of T2 effect on a signal S. Additionally, it selects combinations of TR values and corresponding TI values to make signal from cerebral hemorrhage appear as high signal, and signals from cerebral infarct tissue and cerebral parenchyma appear as isointense or low signal. This allows for rapid and accurate determination of hemorrhagic stroke and, by utilizing low-field magnetic resonance, enhances the accessibility for patients.

The present application discloses an MRI mismatch scoring method, device, equipment, medium and product, relating to the field of intelligent healthcare. The method includes: registering the low-field magnetic resonance stroke discrimination sequence images onto the DWI images, applying the lesion segmentation result on the DWI images to the registered low-field magnetic resonance stroke discrimination sequence images to obtain the region corresponding to the DWI high signal on the registered low-field magnetic resonance stroke discrimination sequence images, calculating the mean image intensity of this region, and comparing the mean image intensity of this region with the mean intensity within the whole brain of the registered low-field magnetic resonance stroke discrimination sequence images to obtain the mismatch ratio. The present application can perform DWI-low-field magnetic resonance stroke discrimination sequence mismatch recognition based on DWI and the low-field magnetic resonance stroke discrimination sequence, so as to improve the efficiency of MRI mismatch scoring and the accuracy of MRI mismatch scoring results.

The invention relates to the technical field of uterus image segmentation, in particular to a woman uterus disease auxiliary diagnosis and evaluation system based on magnetic resonance imaging. After a uterus image of a patient is obtained and a uterine cavity area is positioned, clustering clusters are divided according to pixel gray values by using a clustering algorithm to correspond to different uterus areas. And generating an endometrial region descriptor and screening out an endometrial region by combining the pixel number distribution of each region, the gray features and the shape difference between the endometrial region descriptor and the uterine cavity. Performing gray scale segmentation on the region to extract a high-signal region, and calculating an internal descriptor according to internal edge distribution and morphological characteristics of the high-signal region; meanwhile, an external descriptor is generated based on edge pixel neighborhood gray scale difference. And finally fusing internal and external descriptors to realize auxiliary diagnosis of uterine diseases. According to the method, the uterus image with obvious uterine cancer staging development tendency can be screened out, and a more accurate diagnosis effect is provided for related personnel.

The invention discloses a brain venation plexus vesicle segmentation method, and the method comprises the steps: firstly reinforcing the signal contrast of the whole venation plexus through a high-resolution T1 enhanced image, and then achieving the stable segmentation of vesicles through employing the two steps: the whole segmentation of the venation plexus and the segmentation of the vesicles. In the vesicle segmentation process, as the vesicles are filled with cerebrospinal fluid, the difference between a venation plexus entity (high signal) and the cerebrospinal fluid (low signal) in a T1 enhanced image is utilized, a vesicle segmentation reference threshold value is set, and vesicle voxel separation is realized through low-pass filtering. According to the method, the vesicle structure of the cerebral venation plexus can be segmented, the volume of the vesicle structure can be calculated, and the volume of the vesicle of the cerebral venation plexus can be used as an image marker for auxiliary diagnosis of nerve diseases such as Alzheimer's disease.

This application relates to a magnetic resonance imaging method and a storage medium. The method includes: obtaining a plurality of sampling trajectories of echo signals in the k-space; the plurality of sampling trajectories includes at least two adjacent non-Cartesian sampling trajectories along the phase encoding direction; collecting magnetic resonance signals according to the plurality of sampling trajectories and filling the k-space with the magnetic resonance signals; performing image reconstruction based on the data filled in the k-space to obtain a magnetic resonance image. Using this method can improve the signal acquisition speed and the magnetic resonance imaging speed.

The present invention relates to the technical field of real-time monitoring, and specifically to a real-time monitoring method and system for the processing of granular drugs, including the following steps: Based on the photoacoustic signal data during the processing of granular drugs, a fixed time interval is set to capture the photoacoustic signals of granular drugs under the conditions of thermally induced photoacoustic excitation and pressure-induced photoacoustic excitation, record the amplitudes of the photoacoustic signals, and obtain a photoacoustic signal feature data set. In the present invention, through the capture of photoacoustic signals at fixed time intervals, continuous monitoring of the formation process of granular drugs is realized. By combining thermally induced and pressure-induced photoacoustic excitation, the signal acquisition coverage is improved, the contrast of photoacoustic holographic images is adjusted, the recognition accuracy of trace components is enhanced, the peak shift rate is analyzed, the noise baseline fluctuation is compared, abnormal time points are accurately screened, the accuracy of interface change detection is improved, the particle trajectory data is obtained by combining high-frame-rate imaging, the flow field disturbance distribution is analyzed, the precise control of the processing process is realized, and the stability of the particle quality is ensured.

The present invention relates to a stroke differentiation sequence using low-field magnetic resonance imaging, and a device with the differentiation sequence. An inversion recovery sequence is used, and a TE uses the minimum or near-minimum TE value achievable by a system, thereby minimizing the impact of T2 effect on a signal S; moreover, a combination of a TR value and a corresponding TI value is selected, such that the signal of intracerebral hemorrhage is hyperintense, and the signals of cerebral infarct tissue and brain parenchyma are isointense or hypointense, thereby quickly and accurately identifying hemorrhagic stroke; and low-field magnetic resonance imaging is used, thereby achieving higher accessibility for patients.

The present invention relates to a method and a system for the segmentation of white matter hyperintensities (WMHs) present in magnetic resonance brain images, comprising: providing an array of trained convolutional neural networks (CNNs) with a magnetic resonance brain image; determining, for each of the CNNs and for each voxel, the probability that the given voxel corresponds to a pathological hyperintensity; calculating the average of all the probabilities determined for each voxel; comparing the averaged probabilities for each voxel with a threshold; generating an image mask with the voxels that exceed the threshold.

The invention relates to the technical field of brain image recognition of a stroke patient, in particular to a high-precision recognition method for an MRI image of the stroke patient. Segmenting a brain MRI high-signal region of the stroke patient, and selecting a reference region; calculating the gray difference between the edge reference region and the surrounding region to obtain a diffusion extension coefficient, and screening out an edge reference region and a diffusion reference region; and obtaining an edge strength coefficient according to the pixel position and the gray difference between the pixel position and the two regions, and screening pixel points for fitting to obtain an interception region. The method is based on DWIamp; aDC feature detection intercepts a core area of the area, and the disease degree is obtained according to the gray level of the core area, the edge gradient and the gray level difference of pixel points at the same position of the DWI / ADC. And screening a high-disease-degree region as a cerebral apoplexy focus, and finally realizing automatic identification of the brain MRI image according to focus distribution. According to the invention, the fuzzy boundary of the focus area can be accurately distinguished, so that related personnel can identify the focus area of cerebral apoplexy.

The invention discloses a CSVD cognitive impairment prediction method and a CSVD cognitive impairment prediction system based on Transform and domain self-adaptive T2-FLAIR white matter high-signal radiomics characteristics, and belongs to the technical field of medical image processing and auxiliary evaluation. The system comprises a radiomics feature extraction module, a Transform prediction model module, a domain adaptive training module and an interpretability module. According to the method, image omics features of a white matter high-signal area in a T2-FLAIR image are extracted through a full-automatic process, a context relation between the features is modeled through a Transform architecture, the multi-center data generalization ability is improved in combination with a domain self-adaptive mechanism, feature importance visualization is achieved through Grad-CAM, and the method is used for assisting in predicting cognitive disorder risks and related scores of CSVD patients. According to the method, the problems of strong subjectivity, weak generalization ability, poor interpretability and the like in the prior art are solved, and an efficient tool is provided for early risk assessment of CSVD related cognitive impairment.

The present invention relates to the technical field of lesion area localization in perianal MRI images, and specifically relates to an intelligent localization and annotation method for anorectal lesion areas. The present invention analyzes and obtains high-signal areas in perianal MRI images; according to the gray-scale differences in different areas, as well as the number of pixel points and gray-scale distribution in the areas, the perianal characterization degree is obtained, and then the initial areas are screened out; according to the position distribution and gradient distribution of pixel points on the texture curve in the initial areas, the direction extension coefficient is obtained; combining the gray-scale difference and distance difference between each pixel point in the area and its surroundings, the anorectal lesion localization coefficient is obtained; and then the anorectal lesion area is localized. The present invention can accurately identify the anorectal lesion areas in perianal MRI images, enabling relevant personnel to better judge the situation of anorectal lesions, thereby reducing the probability of misdiagnosis.

The invention relates to a magnetic resonance imaging method based on pseudo-continuous arterial spin labeling, a medium and equipment. According to the magnetic resonance imaging method, on the basis of pseudo-continuous artery spin labeling, PCASL sequence parameters are optimized, then a multi-modal structure image is combined with a PCASL image, a heart artery blood perfusion weighted image is generated, and more accurate parameter estimation is obtained. Furthermore, a proper range is selected to optimize the acquisition sequence through parameter fitting, and the processing process of combining the acquired image and the PCASL image is optimized, so that motion artifacts and physiological noise can be effectively inhibited, the signal-to-noise ratio is improved, and accurate parameters can be acquired more quickly and efficiently. The cardiac artery blood perfusion weighted image generated by the method has various clinical applications such as coronary heart disease diagnosis, myocardial infarction evaluation, cardiac surgery evaluation, cardiac rehabilitation evaluation and the like.

The present invention belongs to the field of magnetic particle imaging, and specifically relates to a magnetic particle imaging time-frequency domain signal filtering method, system, and device based on multi-scale feature fusion, aiming to solve the problem that the existing technology cannot effectively eliminate noise in MPI, resulting in poor magnetic particle imaging quality. The method of the present invention comprises: collecting the original frequency domain signal in the magnetic particle imaging system as the input signal; filtering the input signal through a pre-constructed multi-scale feature filtering neural network model to obtain a filtered time domain signal; the multi-scale feature filtering neural network model includes four filtering submodules; the filtering submodules are constructed based on a sequentially connected convolution layer, a feature filtering structure, and two fully connected layers. The present invention can extract particle and noise-related features in the signal, filter out noise signals, improve the signal-to-noise ratio, and enhance the quality of magnetic particle imaging.

Systems and methods for AI-powered histological fingerprinting in magnetic resonance imaging. MR signal data of an object is acquired using a high sensitivity scanner. Ground truth tissue microstructure data is acquired for the object. A forward model is learned using machine learning. The forward model is used to generate a dictionary or to train a model to map the signals to the histological parameters including the tissue microstructure of a scanner object. A signal-to-signal translation model is also provided to provide signals with improved sensitivity.

This invention provides a multi-channel imaging tube and its preparation method, belonging to the field of medical device technology. The imaging tube includes a tubular body with a cavity for accommodating imaging material. The tubular body has a first central channel and several peripheral channels, with the central channel's axis coinciding with the axis of the tubular body. The peripheral channels are located around the first central channel, and their axes are aligned along the length of the tubular body. Multiple first partitions are arranged in parallel within the tubular body, each partition along its length; a first connecting gap is provided between the edge of each partition and the inner wall of the tubular body. This imaging tube, by filling the tubular body with imaging material to form a continuous three-dimensional imaging network, allows the applicator to be completely surrounded by the high-signal imaging material, enabling accurate identification of the applicator's location during MR scanning and providing a reliable basis for precise execution of treatment plans.

The present invention relates to the technical field of magnetic resonance imaging, and discloses a magnetic resonance diffusion weighted imaging method and a magnetic resonance imaging system for infants and young children, including the following specific steps: using a slice selection gradient pulse to perform slice selection on the imaging tissue; applying an IR fat suppression radio frequency pulse and a slice selection gradient pulse Gs1, and when the signal of the adipose tissue recovers to zero, applying a 90° excitation pulse and a slice selection gradient pulse Gs2, at this time the tissue has the strongest signal intensity; this magnetic resonance diffusion weighted imaging method and the magnetic resonance imaging system for infants and young children, by changing the direction of the 180° slice selection gradient to be opposite to the direction of the 90° pulse slice selection gradient, thereby effectively reducing the artifacts caused by chemical shift, and adding an IR fat suppression radio frequency pulse on the basis of the existing DWI sequence, can automatically adjust the IR pulse according to the signal difference between the fat and other tissues of different patients, and then achieve effective suppression of the image motion artifacts caused by the high signal of the fat.

The present application relates to a coil assembly and a manufacturing method thereof, and a magnetic resonance imaging device, the magnetic resonance imaging device comprising the coil assembly, the coil assembly comprising a substrate and at least one coil array unit, the coil array unit being arranged on the substrate, each coil array unit comprising at least one quadrature coil and a set of linear coils, the set of linear coils comprising at least two linear coils arranged in sequence along a first direction, the at least two linear coils overlapping to form an overlapping area, and the at least one quadrature coil in each coil array unit being arranged one-to-one with the overlapping area. By arranging the quadrature coil, one quadrature coil is arranged in each overlapping area, and the signal-to-noise ratio of the overlapping area can be improved by the quadrature coil, so that the signal receiving quality is improved.

The invention relates to the technical field of lesion area positioning of crissum MRI images, in particular to an intelligent positioning and labeling method for an anorectal lesion area. According to the method, a high-signal area in a crissum MRI image is obtained through analysis; the crissum characterization degree is obtained according to the gray scale difference of different areas and the number and gray scale distribution of pixel points in the areas, and then an initial area is screened out; obtaining a direction extension coefficient according to pixel point position distribution and gradient distribution on the texture curve in the initial region; obtaining an anorectal lesion positioning coefficient by combining the gray difference and the distance difference between each pixel point in the region and the surrounding; and the anorectal lesion area is positioned. According to the method, the anorectal lesion area in the crissum MRI image can be accurately identified, so that related personnel can better judge the anorectal lesion condition, and the misdiagnosis probability is reduced.

The present invention provides an OCT-based attenuation imaging method, which limits the calculation of OAC within a local area of the same type of medium, avoiding the influence on the OAC result caused by the differences in BsAR of different media. When calculating the OAC of a complex sample with multiple layers of media using the traditional DR method, deviations will occur. This is mainly manifested in that when there is an ultra-high or ultra-low signal in the deep layer of the medium, it will affect the OAC of the shallow layer of the medium. If there is an ultra-high signal in the deep layer, the OAC result of the shallow layer will be on the low side; on the contrary, if there is an ultra-low signal in the deep layer, the OAC result of the shallow layer will be on the high side. The method of the present invention solves the problem of estimating the remaining light intensity based on local data by limiting the calculation of OAC within a local area containing only the same type of medium. Thus, the calculation of OAC is not interfered by other layers of media, and the calculation result is more accurate.