38 results about "T2 weighted" patented technology

The invention relates to the technical field of medical image auxiliary diagnosis, in particular to a nuclear magnetic resonance image auxiliary diagnosis method and system based on multi-modal fusion. The method comprises the following steps: step 1, synchronously acquiring a three-dimensional T1 weighted structure image, a T2 weighted fluid attenuation inversion recovery image and diffusion weighted imaging data of a subject, carrying out spatial registration by taking the T1 weighted image as a reference, and executing skull stripping and gray scale standardization; 2, individualized brain region segmentation is carried out based on a brain anatomical map, the lesion sensitivity weight of each modal is calculated for each segmented brain region, and the weight is obtained by quantifying the following parameters; step 3, extracting multi-modal image features in each brain region; and 4, inputting the fusion features of the whole brain region into a multi-task classifier. The standardization and alignment of the multi-mode MRI image in the space and gray level are realized, and the problems of space mismatch and feature interference among different modes are effectively solved.

The invention discloses a bladder cancer level evaluation method and device based on radiomics and a storage medium. The bladder cancer level evaluation method and device are used for improving bladder cancer level evaluation accuracy and reliability. Obtaining a multi-parameter MRI image of the bladder cancer, wherein the multi-parameter MRI image comprises a T2 weighted image, a diffusion weighted image and a dynamic contrast enhanced image; performing tumor focus part sketching on the multi-parameter MRI image; respectively carrying out feature extraction on the delineated T2 weighted image, diffusion weighted image and dynamic contrast enhanced image by using a radiomics package to generate a bladder cancer feature set; performing feature difference screening on the bladder cancer feature set according to the clinical diagnosis tag; carrying out dimension reduction processing on the screened bladder cancer feature set; constructing a single-parameter MRI classification model and a multi-parameter joint classification model according to the bladder cancer feature set after dimension reduction; and using the single-parameter MRI classification model and the multi-parameter joint classification model to jointly evaluate the bladder cancer level of the to-be-detected data.

The invention discloses an MRI (Magnetic Resonance Imaging) image registration generation method based on a Cycle consistency framework, which comprises the following steps of: inputting a T1 weighted image into an image generation network to obtain a T2 weighted image, inputting the T2 weighted image generated by the image generation network and an original T2 weighted image into an image registration network to obtain an aligned T2 weighted image, and outputting the aligned T2 weighted image to a Cycle consistency framework; and inputting the aligned T2 weighted image into an image generation network to obtain a T1 weighted image, and inputting the T1 weighted image generated by the image generation network and the original T1 weighted image into an image registration network to obtain an aligned T1 weighted image. According to the invention, the defects in the prior art can be improved, and the generation quality of the MRI image is improved.

The invention discloses an automatic hepatic fibrosis staging method and system based on deep learning, and the method comprises the following steps: S1, obtaining T1WI and T2WI original image data of a patient, carrying out the data preprocessing and liver segmentation of the original image data, and obtaining a T1 weighted image data set and a T2 weighted image data set; s2, a DMF-Vheat hepatic fibrosis staging model is constructed, the staging model adopts a double-sequence classification architecture and comprises a T1 branch network, a T2 branch network and a fusion network, and a heat conduction operator layer and a mixed attention module are adopted to extract and process complementary information of T1WI and T2WI sequences; s3, putting the T1 weighted image data set and the T2 weighted image data set into the staging model for training, and optimizing the staging model; and S4, outputting a hepatic fibrosis staging result through the staging model. According to the method, complementary information of T1WI and T2WI is fully utilized through the multi-mode deep learning network, and the accuracy and practicability of hepatic fibrosis staging are remarkably improved.

The present disclosure relates to a lumbar vertebra lesion identification method, device, medium and electronic equipment, wherein the method comprises: determining a lumbar vertebra image of a user, the lumbar vertebra image being a T2 weighted magnetic resonance image; inputting the lumbar vertebra image into a target detection model to obtain a lesion category of the lumbar vertebra of the user, wherein the target detection model comprises a down-sampling layer and a pooling layer, the down-sampling layer is used to extract feature data of the lumbar vertebra image, and the pooling layer is used to classify the feature data to obtain the lumbar vertebra lesion category of the user. Compared with the way of judging the lumbar vertebra lesion by artificial judgment and identifying the lumbar vertebra lesion by machine learning method in the related art, the error of identifying the lumbar vertebra lesion type can be reduced, and thus the accuracy of identifying the lumbar vertebra lesion can be improved.

The invention discloses a neurodegenerative change assessment method and device based on nerve melanin MRI, and relates to the technical field of medical imaging and neuroscience assessment, and the method comprises the steps: obtaining MRI image data of a to-be-assessed individual; the MRI image data comprises multi-sequence MRI image data including three-dimensional T1 weighted imaging, two-dimensional T2 weighted imaging and magnetization transfer imaging; preprocessing the MRI image data to obtain preprocessed MRI image data; extracting a region of interest from the preprocessed MRI image data to obtain the region of interest; performing MRI signal feature analysis on the region of interest to obtain MRI signal features; and inputting the MRI signal features into a neurodegenerative change evaluation model, and outputting a neurodegenerative change evaluation result of the individual to be evaluated. According to the application, the non-invasive and accurate evaluation of the neurodegenerative change can be realized by using the unique signal characteristics of the nerve melanin in MRI imaging.

The invention relates to a method and a device for automatically identifying benign and malignant kidney cystic lesions based on magnetic resonance images. The method comprises the following steps: S1, preprocessing and standardizing a T2 weighted image, a diffusion weighted image, an apparent diffusion coefficient image, a T1 weighted image, a skin medullary phase image, a parenchyma phase image and an excretion phase image; s2, respectively training automatic segmentation models corresponding to different images in a targeted manner, and predicting a focus by using the automatic segmentation models; s3, extracting morphological features, first-order features and textural features of the lesions from all the lesions, wherein the morphological features, the first-order features and the textural features comprise features of capsule walls, partitions and nodules of the lesions; screening the extracted features, and constructing a classification model by using the features with good robustness; and S4, preprocessing and standardizing the image of the current patient, respectively inputting the image into each corresponding automatic segmentation model, and operating the segmentation model and the classification model to realize benign and malignant recognition based on image recognition. According to the invention, integrated and automatic benign and malignant accurate diagnosis of kidney cystic lesions is realized.

The invention relates to the technical field of medical image analysis and molecular diagnosis. The invention discloses a multi-mode prognosis evaluation method for local advanced cervical cancer. The method comprises the following steps: acquiring a magnetic resonance T2 weighted image and a tumor tissue sample of the same patient before treatment; based on the magnetic resonance T2 weighted image, a first risk score is output through a deep learning prognosis prediction model, and first risk layering is completed; screening prognosis related key protein combinations based on proteomics expression data of tumor tissue samples to complete second risk stratification or clustering typing; and fusing the first risk layering result and the second risk layering / clustering typing result to generate a comprehensive multi-mode prognostic risk assessment report. According to the method, the MRI image and the proteomics data are fused through the full-automatic deep learning model, non-invasive prognosis evaluation of cervical cancer is realized, high-risk / low-risk patients can be accurately distinguished without manual sketching, individualized treatment decision is supported, and the clinical transformation value is remarkably improved.

The invention relates to the technical field of biomedical nanomaterials, and particularly discloses a chemiluminescence afterglow nanoprobe based on magnetic regulation and control as well as a preparation method and application of the chemiluminescence afterglow nanoprobe. The nanoprobe takes zinc-doped ferroferric oxide (ZnFe2O4, ZF) magnetic nanoparticles with enzyme-like catalytic activity as an inner core, a chemiluminescent molecule QM-CF with a singlet oxygen response characteristic is wrapped, and the nanoprobe is formed by packaging an amphiphilic material DSPE-PEG-2000. According to the nanoprobe, under the action of an alternating magnetic field, a ZF core can be catalyzed to generate singlet oxygen, so that QM-CF molecules are efficiently activated to generate chemical afterglow luminescence with extremely high intensity and long half-life period; and meanwhile, the ZF kernel endows excellent T2 weighted magnetic resonance imaging capability and magnetocaloric effect. According to the invention, remote and controllable activation and enhancement of chemical afterglow luminescence through an external magnetic field are realized, a multi-mode imaging function integrating magnetic resonance imaging, magnetic afterglow optical imaging and thermal imaging is successfully integrated, and a novel tool is provided for precise diagnosis and imaging guide treatment of solid tumors.

A composition that includes a stem cell and a coated iron oxide nanoparticle. The coated iron oxide nanoparticle, being present in the cytoplasm of the stem cell, contains a superparamagnetic iron oxide core that is coated with one or more biocompatible polymers, each of which has a polyethylene glycol group, a silane group, and a linker covalently linking the polyethylene glycol group and the silane group. Also provided is a method for treating an inflammatory disorder in which stem cells are cultured in the presence of coated iron oxide nanoparticles and the cultured stem cells are administered to a subject suffering from an inflammatory disorder. Further disclosed is a method for tracking stem cells in vivo by labeling stem cells with coated iron oxide nanoparticles, administering the labeled stem cell to an individual, and obtaining one or more T2 weighted magnetic resonance images of the individual to track the stem cells.

The invention relates to the technical field of machine learning, in particular to a knee osteoarthritis assessment method based on machine learning, which comprises the following steps: acquiring knee joint T1 and T2 weighted images of a patient with knee osteoarthritis; manually segmenting the knee joint cartilage in the T1 weighted image and measuring the volume of the knee joint cartilage; cutting the T2 weighted image to obtain a meniscus image, manually segmenting the meniscus image to obtain annotation data, and constructing an image data set; training a meniscus automatic segmentation model, and segmenting the meniscus in the meniscus image; carrying out three-classification on meniscus pixels of the middle five layers of the meniscus image, and calculating a meniscus space specificity signal index; the performance of the meniscus space specific signal index in diagnosis of knee osteoarthritis is systematically evaluated. According to the knee joint meniscus damage diagnosis method, the knee joint T1 and T2 weighted images of a patient with knee osteoarthritis are collected, the meniscus automatic segmentation model and a meniscus space specificity signal index calculation method are applied, the damage condition of the knee joint meniscus is accurately evaluated, and knee osteoarthritis diagnosis is achieved. The process can provide a reliable diagnosis basis for orthopedists, so that clinical diagnosis of knee osteoarthritis is effectively assisted.

The application provides an undersampling magnetic resonance image reconstruction method based on a reference image and data correction, comprising the following steps: acquiring a full sampling T1 weighted image as a reference image I ref ; acquiring a full sampling T2 weighted image I T2 , and converting it into undersampling k-space data y; initializing an image sequence I s and initializing a mutual information value sequence MI s ; using a convolutional neural network based on a residual module, establishing an undersampling magnetic resonance image reconstruction model based on the reference image I ref and the undersampling k-space data y; training the undersampling magnetic resonance image reconstruction model; recording the reconstructed image in the reconstruction process and the mutual information value thereof with the reference image; selecting the best reconstructed image based on the mutual information value; and performing iterative k-space data correction on the best reconstructed image to obtain a final reconstructed image. The application can improve the accuracy of the reconstructed image.

The invention relates to the technical field of medical treatment of cesarean section scar pregnancy evaluation and treatment, in particular to a combined evaluation method and system for cesarean section scar pregnancy. According to the method, serum beta-hCG examination, ultrasonic examination and nuclear magnetic examination are completed within 24 hours before an operation, the blood flow type around a gestational sac is measured through ultrasonic, and the blood flow type around the gestational sac is measured; the size of a gestational sac, the size of a gestational sac in a diverticulum and the thickness of a cesarean delivery scar are measured on the T2 weighted sagittal view image through nuclear magnetism, a comprehensive score is preset according to 0-2 scores by combining a serum beta-hCG result (the total score is 10), and surgical treatment is guided according to the score; according to the method, evaluation is accurate and sufficient, operation is easy, convenient and rapid, damage is small, predicted AUC values for operation failure, serious bleeding and continuous existence of lesions are 0.922, 0.955 and 0.862 respectively, applicability and traceability are high, and clinical treatment risks can be effectively guided to be reduced.

The application provides an R2* image synthesis method based on a generative adversarial network, comprising the following steps: S1, data preprocessing: calculating and correcting R2* values from MEGRE sequences; S2, ROI segmentation: after matching the R2* graph with the AALv3 template, the average R2* values of the regions of interest are extracted from the synthesis graph and the real graph; S3, GAN model: the generator inputs the T1 weighted image and the T2 weighted image, and generates the corresponding R2* image; the discriminator distinguishes the synthesized R2* image generated by the generator from the real R2* image obtained from the real MEGRE sequence; S4, quantitative evaluation of the generated image: the normalized mean square error, the peak signal-to-noise ratio and the structural similarity index are used to evaluate the similarity between the synthesized image and the real R2* graph; S5, statistical analysis: the R2* values of the regions of interest in the synthesized graph and the real R2* graph are analyzed and compared. The application realizes the auxiliary diagnosis of neurodegenerative diseases similar to Parkinson's disease from the perspective of medical image synthesis and processing.

The invention discloses a small kidney tumor preoperative diagnosis method, system, medium and device based on MRI, and belongs to the field of medical imaging diagnos.The method comprises the steps that clinical feature data and original MR I image data of a patient to be diagnosed and MRI interpretation feature data obtained through interpretation based on the original MRI image data under the blind method condition are obtained; performing tumor region segmentation in the original MRI image data to generate tumor three-dimensional volume-of-interest data; extracting a radiomics feature set from a non-fat suppression T2 weighted imaging sequence and a contrast enhancement T1 weighted imaging sequence in the original MRI image data based on the tumor three-dimensional volume-of-interest data, and performing weighted fusion calculation on each feature in the radiomics feature set to obtain a radiomics score; and inputting the clinical feature data, the MR I interpretation feature data and the radiomics score into a preset composite diagnosis model for diagnosis, and outputting a diagnosis result. By implementing the method, the accuracy of benign and malignant identification before the small kidney tumor operation can be improved.

The application discloses an adult diffuse glioma machine learning classification system based on image features, relates to the technical field of medical imaging, and has the technical scheme as follows: the system is realized based on medical magnetic resonance imaging (MRI) basic sequence image features, the MRI basic sequence includes a transverse T1 weighted image and a transverse T2 weighted image, and the system comprises an interested region drawing module, an image preprocessing module, a feature extraction module, a feature data processing module, a model training module and a model effect evaluation module. The system realizes classification of adult diffuse glioma based on image features under the framework of the WHO classification guide in 2021.

The application discloses a brain glioma recognition device based on a 5T multi-nuclear magnetic resonance imaging system, which comprises an amide proton imaging acquisition module for receiving a chemical exchange saturation transfer spectrum of amide protons and amide proton concentration; 1 H T2 weighted MRI images, 23 Na MRI images and total sodium concentration maps 23 Na MRI image acquisition module; for receiving 31 P MRS images and phosphorus-containing metabolite concentration of brain regions 31 P MRS image acquisition module; and a recognition module comprising a discrimination network and a segmentation network. The application inputs a sample group into the discrimination network and the segmentation network of the cross-modal data brain tumor, can recognize abnormal signals of different voxels of brain regions, and can outline the boundary of the brain glioma.

The application provides a medical image cross-modality generation method based on double-contrast frequency domain decomposition and adaptive diffusion, which comprises the following steps: acquiring MRI samples and preprocessing, constructing a k-space decomposition and enhancement process, constructing a k-space double-flow adaptive degradation network, and constructing a k-space double-flow adaptive prediction network. Through data enhancement technology and an adaptive Gaussian filter set, different frequency components are separated and enhanced, common edge features are highlighted, and the quality of generated information is improved. By using a residual noise diffusion model, high-resolution images with rich details and high diversity are generated while maintaining smooth areas and large-scale structures of images. By introducing anatomical structure consistency constraints and multi-modality fusion optimization strategies, the complementary information of T1 and T2 weighted MRI is fully utilized, and the generation effect and visual authenticity of the medical image cross-modality generation in a complex scene are significantly improved.

The application discloses an infant brain T2 weighted magnetic resonance imaging optimization method based on a fast spin echo sequence. First, T1, T2 and PD quantitative imaging of the infant brain from 0 to 24 months old is collected to obtain T1, T2 and PD values of the infant brain white matter and gray matter regions, and according to the relationship characteristics of the infant brain white matter T2 value and the gray matter T2 value, the infant is divided into different month groups. Then, based on the 3D T2 weighted imaging of the variable flip angle fast spin echo sequence, the signal intensity of the infant brain white matter and gray matter under different refocusing flip angle chains is calculated through an extended phase graph algorithm, and the best flip angle chain design scheme of each group is determined with the maximum white matter / gray matter contrast as the target. The application fills the blank of the infant brain T2 weighted imaging optimization, formulates the best flip angle chain optimization scheme of different month groups, and thus significantly improves the contrast of the infant brain T2 weighted imaging.

The invention relates to the technical field of magnetic resonance imaging, and discloses a thin-layer oblique coronal focusing view diffusion weighted imaging method of a magnetic resonance system, which comprises the following steps of: extracting a three-dimensional space walking path of a target bile duct based on magnetic resonance pancreaticobiliary duct water imaging and a cross section T2 weighted image, performing optimal plane fitting, and determining a personalized oblique coronal scanning plane; in the plane, diffusion weighted imaging is carried out by adopting a small visual field set according to bile duct projection, a layer thickness not higher than 3.0 mm and a low b value not higher than 100 seconds / square millimeters and a high b value not lower than 800 seconds / square millimeters, and echo time is set to a minimum value allowed by a system. Through cooperation of personalized plane positioning, thin-layer high-b-value imaging and intelligent quality control, the problems that conventional magnetic resonance imaging is incomplete in display of an extrahepatic cholangiocarcinoma lesion anatomical structure, low in signal-to-noise ratio, multiple in motion artifacts and difficult to accurately evaluate the longitudinal infiltration range of a tumor are solved.

The invention relates to a lumbar vertebra lesion recognition method and device, a medium and electronic equipment, and the method comprises the steps: determining a lumbar vertebra image of a user, and the lumbar vertebra image is a T2 weighted nuclear magnetic resonance image; the lumbar vertebra image is input into a target detection model, the lesion category of the lumbar vertebra of the user is obtained, the target detection model comprises a down-sampling layer and a pooling layer, the down-sampling layer is used for extracting feature data of the lumbar vertebra image, the pooling layer is used for carrying out classification processing on the feature data, and the pooling layer is used for obtaining the lesion category of the lumbar vertebra of the user; and obtaining the lumbar vertebra lesion category of the user. Compared with a mode of manually judging the lumbar vertebra lesion and identifying the lumbar vertebra lesion through a machine learning method in the prior art, the method can reduce errors of identifying the type of the lumbar vertebra lesion, and can improve the accuracy of identifying the lumbar vertebra lesion.

The invention provides a Douglas fossa occlusion prediction method based on multi-mode MRI (Magnetic Resonance Imaging) feature fusion. The method comprises the following steps: collecting multi-modal 3D MRI data of a patient, wherein the multi-modal 3D MRI data comprises pelvic cavity 3D T1 weighted and T2 weighted MRI image data; the method comprises the following steps of: firstly, carrying out multi-modal MRI (Magnetic Resonance Imaging), preprocessing the multi-modal MRI to realize image registration, mapping the multi-modal MRI to the same space, and secondly, realizing local contrast enhancement and marginal definition optimization of the MRI image through adaptive histogram equalization; secondly, designing a parallel-based double-branch 3D-Mama network to extract long-term spatial dependence and continuous features of the weighted MRI images T1 and T2; a 3D channel and space double attention fusion module is further designed, and multi-mode MRI feature deep fusion and adaptive feature selection are achieved; and finally, mapping the learned multi-modal MRI fusion features to a classification space by using a multi-layer perceptron, and carrying out POD occlusion classification prediction.

The invention provides a medical image processing method and system, computer equipment and a storage medium, and belongs to the field of image processing.The medical image processing method comprises the steps that T1 weighted imaging and T2 weighted imaging in brain and neck medical images are collected, multi-scale feature extraction is conducted on the T1 weighted imaging and the T2 weighted imaging, and brain and neck region features are obtained; generating a first target image according to the brain and neck region features; generating a second target image from the first target image based on a dense connection mechanism; optimizing a signal of a focus area in the second target image through a focus perception loss function fused with the focus automatic segmentation result to obtain an optimized target image; and synthesizing a cross-modal brain target image and a cross-modal neck target image according to the first target image, the second target image and the optimized target image. According to the method, the problem of key sequence deletion in multiple sclerosis diagnosis is solved, the retention rate of small focuses is increased, and reliable technical support is provided for MS early screening.

The application provides a renal artery image registration method and device, electronic equipment and storage medium, and relates to the technical field of image processing. The method comprises: processing a magnetic resonance T2 weighted coronal sequence to determine a key frame image containing a spine structure; splicing at least two frames of images in a digital subtraction angiography sequence to generate a reference image containing a spine panorama; back-projecting head-to-foot direction coordinates of a target slice center in a patient coordinate system in magnetic resonance angiography to a corresponding pixel row in the key frame image to obtain first pixel coordinates; calculating a physical displacement corresponding to the first pixel coordinates from a starting pixel coordinate; calculating a pixel displacement corresponding to the physical displacement in the reference image, and determining a positioning result of a renal artery entrance in the reference image based on the pixel displacement and a mapping point of the starting pixel coordinate in the reference image. Through the method provided by the application, the amount of contrast agent and related risks are effectively reduced.

The invention relates to a local anesthesia dosage control system based on CFD (Computational Fluid Dynamics) and PC-MRI (Personal Computer-Magnetic Resonance Imaging) coupling and a working method thereof, and the system comprises an information collection unit which is used for collecting physical examination and medical evaluation information of a patient; the geometric unit is used for acquiring a T2 weighted MRI image sequence; the calculation model unit is used for determining the range of a lumbar, thoracic and vertebral segment calculation domain; a measurement unit for phase encoding using PC-MRI; the cerebrospinal fluid dynamic unit is used for acquiring an initial flow field of cerebrospinal fluid flow of the spinal canal of the patient; the dosage selection unit is used for selecting the injection dosage of the local anesthetic; the intrathecal administration calculation unit is used for establishing coupling analysis to obtain a local anesthetic concentration value of a blocking plane; and the dose evaluation unit is used for quantitatively evaluating the concentration value of the local anesthetic. The system is beneficial for improving the accuracy and safety of local anesthesia medication.

The invention discloses a magnetic resonance imaging sequence trajectory rewinding compensation method and device, a medium and a product, and relates to the technical field of magnetic resonance imaging, and the method comprises the steps: collecting the actual gradient waveforms of three gradient axes in a set time period after an excitation radio frequency pulse is applied to a water model in a magnetic resonance scanner in the current repetition time; setting ideal gradient waveforms of three gradient axes in a set time period; any gradient axis is determined as a current axis; determining the residual gradient time moment of the current axis based on the actual gradient waveform and the ideal gradient waveform of the current axis in the set time period, and determining the waveform parameter of the current axis of the rewinding compensation gradient pulse; and applying a component of the rewinding compensation gradient pulse on the current axis to the current axis based on the residual gradient time moment of the current axis and the waveform parameter of the current axis of the rewinding compensation gradient pulse at the end moment of the phase disturbing gradient so as to realize the rewinding compensation of the magnetic resonance imaging sequence trajectory. According to the invention, the problems of quantitative deviation and image blurring of T2 weighted imaging are solved.

The invention discloses a virtual magnetic resonance imaging system, which belongs to the technical field of magnetic resonance imaging, and comprises a layer thickness selection module, an interlayer interference module, an excitation angle module, an excitation frequency module, a TR dynamic adjustment module and a TE dynamic adjustment module. The thin layer thickness improves the layer selection direction resolution, and the thick layer thickness improves the signal-to-noise ratio; interlayer cross interference can be reduced by controlling the interlayer spacing, useless signal superposition is avoided, and the signal-to-noise ratio is further improved by combining with increase of the layer thickness; the adaptive excitation angle can optimize the energy consumption and the relaxation time, the energy demand is reduced at a small angle, and the relaxation time is shortened; the signal-to-noise ratio can be quantitatively improved by adjusting the excitation frequency, and the one-time excitation frequency and the signal-to-noise ratio are improved by two times; the signal strength can be guaranteed and the time waste can be avoided by dynamically adjusting the TR; tE is selected for focusing T2 weight, organization difference display and a basic signal-to-noise ratio are considered, and the image quality and the imaging efficiency are integrally optimized.

The invention relates to a noninvasive assessment method for radiation cognitive function impairment, and the method comprises the following steps: S100, obtaining a nuclear magnetic resonance image, and carrying out the standardization of the nuclear magnetic resonance image, and the nuclear magnetic resonance image comprises a T2 weighted image and a BOLD image; s200, after second-order analysis, parameter estimation and multiple comparison are carried out on the standardized T2 weighted image and the BOLD image, data with morphological differences are obtained through data processing, and a T2 weighted structure image and a BOLD functional image are obtained; s300, carrying out VBM analysis based on the T2 weighted structure image, and extracting a gray matter volume value of the differential brain region; s400, performing FC analysis based on the BOLD functional image; and S500, constructing a VBM-FC correlation analysis framework, and forming a joint index and integrated evaluation model.

The invention relates to the technical field of intelligent medicine, in particular to a Crohn disease intestinal fibrosis condition early warning method based on large model driving, which comprises the following steps: acquiring an intestinal MRI image sequence of a Crohn disease patient; inputting the intestinal MRI image sequence into a pre-trained intestinal structure analysis large model, and positioning the position of an initial intestinal fibrosis region; determining whether the intestinal lumen stenosis degree of the patient with the Crohn disease reaches the standard or not based on the intestinal lumen stenosis progress index, and determining a feature comparison threshold value for adjusting the large model according to the difference value; determining whether the acute inflammation shielding effect of the patient with the Crohn disease is qualified or not based on the intestinal wall edema characterization parameters, and determining and adjusting the T2 weighted imaging modal weight of the large model according to the relative difference; and determining whether the fibrosis seepage channel smoothness of the patient with the Crohn disease is qualified or not based on the fibrosis seepage resistivity, and determining and adjusting the boundary loss weight of the large model according to the ratio. The early warning accuracy of the intestinal fibrosis condition of the Crohn's disease is improved.

The invention discloses a method for predicting and identifying a joint replacement postoperative infection type based on a preoperative MRI (Magnetic Resonance Imaging) image, which comprises the following steps that: the preoperative MRI image of a patient, which comprises a T1 weighted sequence and a T2 weighted sequence, is acquired, and the T1 weighted sequence and the T2 weighted sequence both comprise three view angle images of coronal view, sagittal view and transection view to form an independent six-channel input tensor; automatically segmenting a joint infection related structure in the MRI image by introducing an image segmentation model of an attention mechanism, enhancing the response to high-risk infection region features, and generating a pixel-level mask as a subsequently classified ROI (Region of Interest); and inputting the six-channel input tensor into a deep convolutional neural network, guiding feature focusing in combination with an infection risk area mask generated by the image segmentation model, extracting high-order semantic features of an infection risk area, and outputting a three-classification prediction result through a full connection layer. The three-classification prediction result comprises superficial infection, deep infection and non-infection. And outputting a three-classification prediction result.