22 results about "Mri brain" patented technology

The invention discloses a multi-axis RWKV-UNet + + multi-mode MRI (Magnetic Resonance Imaging) brain tumor segmentation method, and belongs to the technical field of medical image processing. According to the invention, multi-modal MRI three-dimensional body data is input and preprocessed, and fusion features are output through a modal fusion module; the fusion features are input into an encoder containing multi-axis RWKV sequence modeling, and long-range dependence is extracted; after the output of the encoder is processed by the bottleneck layer, the global Token aggregator converges the global context and reinjects the global context; the enhanced features are input into a UNet + + nested topology decoder, the jump features are fused with the up-sampling features after being subjected to jump RWKV semantic alignment, and finally a three-dimensional segmentation probability graph is generated through mapping. The method is mainly used for accurate three-dimensional segmentation of the multi-mode MRI brain tumor, and provides support for clinical brain tumor diagnosis and treatment.

The invention discloses an improved multi-modal MRI (Magnetic Resonance Imaging) brain tumor segmentation method, which comprises the following steps of: firstly, acquiring a multi-modal MRI image containing a brain tumor, and preprocessing the multi-modal MRI image; secondly, based on the preprocessed multi-modal MRI image, enhanced modal correlation modeling and feature fusion are carried out, and smooth edge features are obtained; and finally, based on the smooth edge features, performing multi-task decoding and result output to obtain a complete brain tumor segmentation image. According to the method, the key problems of insufficient modal feature alignment, inter-task information segmentation, fuzzy segmentation boundary and the like in the existing multi-modal brain tumor segmentation are effectively solved.

The invention provides a scalp-to-brain magnetic resonance imaging generative model training method, a scalp-to-brain magnetic resonance imaging generative model generating method and scalp-to-brain magnetic resonance imaging generative model generating equipment, and relates to the technical field of image processing. The network comprises an encoder, a generator, a feature discriminator and an image discriminator, the encoder and / or the generator is integrated with a neural network model based on a frequency domain attention mechanism, and the encoder and the generator jointly form a target generation model after being trained; and the model parameters are jointly optimized by using a composite loss function comprising a frequency domain loss item. According to the method, the problems that an existing brain image generation technology neglects scalp structure information and is difficult to model global dependence and high-frequency details can be solved, brain magnetic resonance imaging data can be highly generated from scalp magnetic resonance imaging data, the structural integrity and topological fidelity of the generated data can be improved, and the application reliability of auxiliary diagnosis can be improved.

The invention discloses a multi-parameter MRI brain age prediction method and device based on uncertainty perception, equipment and a medium, and the method comprises the steps: determining a first feature sequence of an sMRI image, a second feature sequence of a DTI image, and a third feature sequence of an fMRI image through a feature extraction module; and determining a GM sampling vector, a DTI sampling vector and an fMRI sampling vector based on the first feature sequence, the second feature sequence and the third feature sequence by using a probability distribution encoder, and performing brain age prediction by using the GM sampling vector, the DTI sampling vector and the fMRI sampling vector. According to the method and the device, the fusion feature vector determined by fusing the sMRI, DTI and fMRI information is utilized to perform brain age prediction, so that multi-dimensional information of the brain structure and function can be more comprehensively acquired, and the accuracy of brain age prediction is improved. Meanwhile, a probability distribution encoder is introduced in the process of fusing the multi-parameter MRI images, multi-parameter MRI feature mapping is converted into Gaussian distribution through the probability distribution encoder to model feature uncertainty, richer potential information in the multi-parameter MRI images is captured, and the accuracy of brain age prediction is further improved.

A method for assessing a patient's brain disease state from brain images includes acquiring the brain images, partitioning them into predefined regions based on anatomical landmarks, extracting disease-indicative features, using a pretrained model to generate a disease-associated biomarker from the features, and determining the patient's brain disease state based on the biomarker.

The application relates to a brain tumor segmentation method and device based on TriSAM-UNet, which comprises the following steps: S1, collecting a multi-modal MRI brain tumor image dataset, and performing pretreatment and data enhancement; S2, constructing a hybrid segmentation network TriSAM-UNet, designing a hybrid module SAHM, capturing three-dimensional dependence and cross-region jump correlation, and realizing all-around global context perception; S3, constructing an FUE module, inhibiting fuzzy and noise response to ensure that high-confidence features are preferentially transmitted; S4, constructing a WFAB module, and restoring the boundary and texture details lost in up-sampling; and S5, training and optimizing the network model and performing full-automatic segmentation on the brain tumor image. The application aims to solve the problems of insufficient global context capture, indiscriminate forwarding of noise features in jump connection and loss of boundary details in up-sampling in brain tumor segmentation, and is suitable for the field of brain tumor image analysis.

The application discloses a 2.5D MRI brain tumor segmentation method based on Mamba state space modeling and belongs to the technical field of medical image segmentation. The application aims at the problems of the existing brain tumor segmentation method, such as long-range dependence modeling deficiency, high calculation cost, limited boundary segmentation precision and poor adaptability of the Mamba architecture to the 2.5D segmentation scene, and a cross-slice priority traversal module suitable for the 2.5D slice space structure is designed, local details and global semantic features are balanced through a feature calibration and controlled fusion module, and a global perception and edge enhancement dual-decoder structure is constructed to decouple tumor global modeling and boundary modeling tasks. The application reduces the calculation complexity while ensuring the segmentation precision, solves the problems of tumor segmentation discontinuity and boundary blur and is suitable for clinical conventional image workstations and has a good application prospect.

The invention relates to a multi-modal MRI brain tumor segmentation method based on convolution attention, and the method comprises the steps: carrying out the parallel coding of multi-modal medical image data, and extracting the specific low-layer space features of each modal; performing feature fusion and high-level coding on the low-level spatial features to obtain shared features; carrying out convolution attention processing on the shared features to obtain enhanced features; and carrying out decoding processing on the enhanced features to obtain a segmentation result. According to the method, more accurate tumor boundary segmentation and tissue differentiation can be realized in medical image analysis such as brain MRI (Magnetic Resonance Imaging), meanwhile, the parameter efficiency, the calculation speed and the generalization ability under limited data of the model are remarkably improved, and the clinical practicability is enhanced.

The invention discloses a model, a method, an electronic device and a medium for MRI brain glioma segmentation, and the model comprises a learnable bias field correction module which is used for adaptively estimating and correcting a low-frequency bias field in an original brain glioma MRI image; the multi-resolution feature extraction module is used for carrying out multi-resolution feature extraction and outputting a multi-resolution feature graph group containing different resolution features and shallow layer features; the multi-resolution feature fusion module is used for carrying out feature fusion on the multi-resolution feature map group and recovering relatively high resolution to obtain deep features; and the statistical space gating fusion module is used for receiving the deep layer features and the shallow layer features, performing feature fusion through channel statistical screening and a space attention gating mechanism, and outputting an MRI brain glioma segmented image. The technical problems that an existing brain glioma MRI image segmentation method is generally poor in adaptability, insufficient in boundary continuity, low in fusion efficiency and poor in global and local feature balance are solved.

The invention relates to a multi-modal brain tumor segmentation method and device based on hierarchical attention enhancement, and the method comprises the following steps: S1, collecting a multi-modal MRI brain tumor image, constructing a data set, and carrying out the preprocessing; s2, constructing a multi-view region sensing feature aggregation module, performing feature extraction and attention enhancement on the brain tumor MRI, and fusing local features and global features; s3, constructing an adaptive correlation collaborative enhancement fusion module, focusing the key region, and carrying out adaptive fusion; s4, network parameters are updated through a category importance balance region perception composite loss function, and a network model is trained and adjusted to optimize segmentation performance; and S5, performing MRI brain tumor segmentation according to the trained segmentation model. According to the method, the multi-modal information can be effectively processed from the MRI image, the relation between different regions of the brain tumor is extracted and fused, and the brain tumor is accurately positioned and accurately segmented by combining the local features and the global features.

The invention provides a method for constructing a multi-sequence MRI (Magnetic Resonance Imaging) tumor diagnosis model, which comprises the following steps of: performing iterative training on a constructed MRI brain tumor diagnosis model: performing first-stage training: performing iterative training on the constructed MRI brain tumor diagnosis model by adopting a Focal loss function based on a first original sample set; after the first-stage training is completed, a difficult case list and an optimal MRI brain tumor diagnosis model are obtained; when the number of iterations of the first-stage training and / or the AUC index reach a corresponding preset threshold value, entering second-stage training; second-stage training: adding the obtained difficult case list to a second original sample set as a second-stage training sample set; based on the second-stage training sample set, performing iterative training on the optimal MRI brain tumor diagnosis model by adopting a joint loss function formed by Focal loss and Dice Loss until the model is converged; structural defects of single-round training and fixed matching are systematically solved, and more balanced optimization of the model among minority class recall, overall distinction degree and deployment threshold performance is achieved.

A brain magnetic resonance imaging (MRI) analysis device and method using a neural network are disclosed. The brain MRI analysis device using a neural network according to one embodiment comprises: a memory for storing a neural network model; and a processor, which is connected to the memory so as to control an analysis device, wherein the processor receives one or more brain MRI images so as to generate input data, and inputs the input data into the neural network model so as to acquire output data, and the neural network model is trained to determine neurological prognosis of a cardiac arrest patient if the input data is input into the neural network model.

The utility model discloses a microwave antenna and a cerebral hemorrhage detection system, and relates to the technical field of biomedical engineering. Wherein the microwave antenna comprises a first curved surface dielectric layer (1a) and a second curved surface dielectric layer (1b); the first curved surface dielectric layer (1a) and the second curved surface dielectric layer (1b) are respectively provided with a first excitation port (36) and a second excitation port (55); wherein the first curved surface dielectric layer (1a) and the second curved surface dielectric layer (1b) are respectively provided with a plurality of groups of first radiation patches connected with the first excitation port (36) and a plurality of groups of second radiation patches connected with the second excitation port (55). The objective of the utility model is to solve at least one of the technical problems of ionizing radiation, high cost, high probability of missing gold first-aid time due to overlong time, unsuitability for bedside detection and pre-hospital first aid and the like in current CT and MRI brain image examination.

This application discloses an automatic hippocampal segmentation method based on HAU-Net and gated spatial attention. The method includes: acquiring a three-dimensional MRI brain image of the target object to be segmented; sampling at least one target slice from the three-dimensional MRI brain image, and constructing a multi-channel two-dimensional input feature map for each target slice, wherein the multi-channel two-dimensional input feature map is formed by stacking multiple consecutive slices containing the target slice along the channel dimension; inputting the multi-channel two-dimensional input feature map into a pre-trained hippocampal segmentation network to output a hippocampal segmentation probability map; and performing thresholding processing on the hippocampal segmentation probability map to obtain the final binarized hippocampal segmentation result. This achieves an optimized balance between segmentation accuracy and computational efficiency, significantly improving the accuracy and continuity of hippocampal segmentation while maintaining the computational efficiency of the two-dimensional network.

The invention discloses an incomplete multi-mode brain tumor segmentation method based on a double-level uncertainty guidance network, and the method achieves the precise positioning of a brain tumor. The method comprises the following steps: S1, carrying out preprocessing and data enhancement on a brain tumor image, and constructing a data set by constructing a random modal missing condition and setting different missing proportions; s2, constructing an uncertainty guidance cross-domain enhancement module, explicitly quantifying uncertainty in a modal inner layer, and cooperating with a spatial domain and a frequency domain to suppress local noise to obtain enhanced single-modal features; s3, constructing an uncertainty guidance confidence coefficient calibration fusion module, and dynamically adjusting a fusion weight on an inter-modal level by using uncertainty difference to obtain a fusion feature; s4, performing MRI brain tumor segmentation according to the trained MRI brain tumor segmentation model; according to the method, the robustness and segmentation precision of the model in a clinical complex and missing scene are remarkably improved by modeling from the intra-modal level and the inter-modal level and suppressing uncertainty.

The invention provides a children MRI brain tumor grading method. The children MRI brain tumor grading method comprises the following steps: preprocessing brain MRI image sequences of T1, T1c, T2, Flair and ADC modals to obtain tensors corresponding to the brain MRI image sequences; inputting the tensor of the T1 mode into an SE feature extraction channel for feature extraction, wherein the SE feature extraction channel performs feature extraction based on an SE attention mechanism; tensors of the T1c, the T2, the Flair and the ADC mode are input into an MSFE multi-scale feature extraction module for feature extraction; features extracted by the SE feature extraction channel and the MSFE multi-scale feature extraction module are subjected to feature fusion and then are input into a classification module for classification prediction to obtain a classification prediction result; according to the method, undetectable subtle changes in early tumors can be recognized, and the calculation load is effectively reduced while high precision is maintained.

The invention provides an MRI (Magnetic Resonance Imaging) brain tumor small focus high-precision detection system based on HMA-DETR (Hidden Markov Amplification-DETR), which is characterized in that an HMA-DETR model is an improved RT-DETR model and specifically comprises a backbone network, a hybrid encoder and a converter encoder; wherein a plurality of hybrid gating convolution modules are arranged on the backbone network; a multi-scale expansion interactive attention module is arranged on the hybrid encoder to replace the AIFI module; the hybrid encoder is also provided with a plurality of spatial attenuation enhancement modules to replace a RepC3 module in the cross-scale feature fusion CCFM. And processing the output characteristics of the hybrid encoder through a converter encoder to obtain a brain tumor small focus detection result. According to the MRI brain tumor small focus high-precision detection system based on the HMA-DETR, the small target detection precision and the cross-modal robustness are remarkably improved while the reasoning speed is kept.

The invention relates to a brain tumor segmentation method and device based on multi-view feature dynamic interactive fusion, and the method comprises the following steps: S1, collecting a multi-modal MRI brain tumor image, constructing a data set, carrying out the preprocessing, and highlighting the details and features of a tumor; s2, constructing a multi-dimensional visual angle dynamic interaction fusion module, and realizing multi-visual angle feature dynamic fusion through three orthogonal visual angles in combination with a local detail and global space relationship; s3, constructing a hierarchical gating adaptive enhancement fusion module, carrying out dynamic weight adjustment, and carrying out selective fusion on cross-scale local details and global semantic information; s4, network parameters are updated through a self-adaptive structure sensing joint loss function, and a network model is trained and adjusted to optimize segmentation performance; and S5, performing MRI brain tumor segmentation according to the trained network model. According to the method, through multi-view information cooperative interaction and enhancement of cross-scale feature fusion, the recognition capability of the brain tumor area is effectively improved, and accurate segmentation of the brain tumor is realized.

The application relates to the technical field of digital image processing, and provides a CT-MRI brain tumor target segmentation method based on cross-modal information interaction, which comprises the following steps: after CT images and MRI images of glioma patients in a GLIS-RT data set are rigidly registered and normalized, the CT images and the MRI images are divided into a training set and a test set; a CI 2 Net network: in the encoding stage, the network respectively extracts modal features of CT and MRI, information fusion between the modes is realized through a CMI module, and a RDT module is used to replace a conventional convolution layer in a bottleneck layer; in the decoding stage, different levels of features are fused through an FAA module to generate features with discrimination ability; in the CI 2 Net network, the decoder of the CI 2 Net network, the decoder of the CI 2 Net network; the CI 2 Net network is used for segmenting a GTV region. The application improves the accuracy of glioma GTV segmentation.

The application discloses a kind of multi-axis RWKV-UNet++ multimodal MRI brain tumor segmentation method, belong to medical image processing technical field.The application is by inputting multimodal MRI three-dimensional body data and pre-processing, and fusion feature is output by mode fusion module;Fusion feature is input to the encoder of the modeling containing multi-axis RWKV sequence extraction long-range dependence;Encoder output is processed by bottleneck layer, and global context is converged by global Token aggregator and is injected back;Enhanced features are input into UNet++ nested topology decoder, and after jump connection RWKV semantic alignment, up-sampling features are fused, and finally three-dimensional segmentation probability map is generated.The method is mainly used for the accurate three-dimensional segmentation of multimodal MRI brain tumor, and provides support for clinical brain tumor diagnosis and treatment.

The invention discloses a 3D MRI brain tumor segmentation method based on a multi-distillation mechanism, and belongs to the field of medical image segmentation. According to the method, a three-dimensional segmentation framework composed of a teacher network and a student network is constructed, and high performance of the teacher network is utilized to guide students to learn on the network. By introducing three complementary mechanisms of feature distillation, student self-distillation and main and auxiliary logic distillation, the feature expression and classification discrimination ability of the student network is enhanced. Wherein the characteristic evolution consistency constraint is introduced into the characteristic distillation to maintain the learning direction consistency, and the main and auxiliary logic distillation is combined with the Logit constraint of the boundary and the internal region and the temperature smoothing Softmax-KL divergence to optimize the structure and category consistency. And finally, introducing a channel and a space attention module into the network, and realizing self-adaptive balance of multiple distillation tasks through dynamic weight distribution.

This invention proposes a three-dimensional MRI brain tumor segmentation method based on a diffusion model, belonging to the field of medical image generation technology. The method involves acquiring MRI images of different modalities of the brain tumor site and their corresponding ground truth values ​​for segmentation; adding noise to the ground truth values ​​for segmentation through forward diffusion to obtain noisy segmentation data; performing iterative discrete wavelet transform on the MRI images to obtain wavelet information at various levels; constructing a brain tumor segmentation model based on introducing a cross-attention mechanism into the U-Net network structure to improve the diffusion model; and training the model using the aforementioned data; sampling random noise from a prior distribution, concatenating it with the MRI image to be segmented, and inputting it into the trained model to obtain the corresponding inference target domain segmentation image. This method can extract multi-scale features from multimodal MRI and significantly reduce the number of model parameters, integrate conditional information into the network, and enhance the feature expression of key regions, thereby improving the accuracy of MRI brain tumor segmentation.