87 results about "Lesion type" patented technology

The invention discloses an intelligent detection method and device for fusing a medical image learning image, and relates to the technical field of medical image processing. The method comprises the following steps: acquiring and preprocessing a bimodal medical image, and extracting a feature map through multi-scale decomposition; constructing a cross-modal correlation model, and setting a modal attention mechanism (embedding anatomical structure prior guidance feature complementation) and a morphological attention mechanism (setting lesion morphological constraint weight); the method comprises the following steps: collecting multiple types of image samples, pairing according to a focus form and an imaging mode to construct a bimodal joint data set, and correlating and labeling to generate a training data set with modal attributes; after a multi-stage iteration training model, inputting the preprocessed image to carry out feature fusion so as to obtain a fused image; and generating a lesion probability graph according to the fused image, positioning a lesion area through multi-threshold segmentation, and outputting a detection result. The system comprises a data acquisition module, a preprocessing module and the like. The method improves the accuracy and reliability of medical image detection, and is suitable for clinical multi-modal image analysis.

The invention relates to the technical field of medical image processing, in particular to a CT image intelligent analysis system for pneumonia auxiliary screening. The method comprises the following steps: firstly, preprocessing a chest CT image and detecting a candidate focus area; secondly, extracting a topological feature, a deep convolution feature and a texture statistical feature based on a persistent coherence theory from each candidate focus, and performing feature fusion through a multi-head self-attention mechanism to generate a unified focus representation vector; mapping the lesion characterization vectors to a pre-constructed radiology knowledge graph, adopting a graph neural network for reasoning, and outputting the pneumonia suspected probability and lesion classification of each lesion; and finally, performing fusion and uncertainty quantification on the analysis results of the plurality of focuses by adopting an evidence theory, and generating a comprehensive screening report. According to the method, complex-form lesions are effectively identified through topological features, accurate identification of lesion types is realized through knowledge graph reasoning, and diagnosis uncertainty quantification is provided through an evidence theory.

The invention discloses a cerebral vessel anomaly analysis method based on a transcranial Doppler ultrasound image, which comprises the steps of TCD image data acquisition, TCD image space-time filtering preprocessing, automatic blood vessel positioning and three-dimensional modeling, the method comprises the steps of generating a standardized blood vessel segmentation model and a static blood vessel network graph, extracting blood flow quantitative parameters and spectrum qualitative features, analyzing multi-blood vessel space correlation features through a graph attention network, performing primary classification on blood flow states, performing secondary classification on lesion types, making risk grading judgment, generating intervention suggestions and outputting a structured report. By setting an automatic parameter measurement method, the manual operation time of doctors is shortened, and the speed and accuracy of lesion judgment are improved through automatic standardization judgment of frequency spectrum forms and association of multi-dimensional parameters. Meanwhile, the experience requirement of doctors for judging the lesion is reduced, and reference is conveniently provided for batch screening of primary hospitals.

The invention relates to the technical field of image processing, in particular to a clinical ultrasonic image auxiliary screening system based on deep learning. The system comprises an ultrasonic image preprocessing module, a lesion area segmentation module, an image lesion marking module and a lesion auxiliary screening module, and can obtain a clinical ultrasonic image and perform graying processing and neighborhood gray level equalization processing to generate a clinical ultrasonic equalization image; performing pixel normalization and tissue boundary fuzzy-based lesion region segmentation on the clinical ultrasonic equalization image to generate clinical ultrasonic image lesion region blocks; carrying out image lesion marking through the clinical ultrasonic image lesion area block to generate a clinical ultrasonic lesion marking image set; and constructing an image lesion auxiliary screening model and carrying out lesion auxiliary screening so as to output lesion positions, lesion types and lesion severity corresponding to lesions on the clinical ultrasonic image. According to the invention, accurate analysis of clinical ultrasonic images can be realized, so that lesion screening can be efficiently completed in an assisted manner.

The invention belongs to the technical field of image detection, and discloses an enteroscopy auxiliary diagnosis method and system based on artificial intelligence, and the method comprises the steps: obtaining enteroscopy image data in an enteroscopy process, and synchronously extracting corresponding auxiliary collection information from the enteroscopy image data; caching the enteroscope image data and the auxiliary acquisition information frame by frame and keeping timestamps aligned; performing standardized correction and adaptive enhancement processing on the enteroscope image data after timestamp alignment; a self-adaptive gamma dynamic enhancement mechanism is introduced, and intestinal tract wrinkles are reserved to obtain an enhanced image; lesion positioning and category screening are conducted on the enhanced image through the target detection model, multi-scale feature aggregation and time sequence consistency constraint are conducted in combination with auxiliary collection information, and lesion candidate areas are generated; performing morphological, texture and boundary structure analysis on the lesion candidate region, and extracting lesion morphological features; and the efficiency and the reliability of enteroscopy diagnosis can be improved.

The invention provides a target detection-based blood vessel feature analysis method and device, equipment and a medium, and the method comprises the steps: carrying out the processing of an input blood vessel image through an initialized feature extraction module, generating a pre-training model, carrying out the feature extraction of the blood vessel image through the pre-training model, and obtaining image feature data; inputting the lesion area into an analysis module, and performing image segmentation on the lesion area through the analysis module to obtain segmented image features; the segmented image features comprise image data of a calcified region and / or a non-calcified region; and respectively inputting the segmented image features into a calcified plaque analysis module and a non-calcified plaque analysis module to obtain the type of a vascular lesion area, further analyzing the density distribution of mixed plaques to determine whether the mixed plaques are in a splicing type or a uniform type, and finally generating an analysis result containing the lesion type and the stenosis degree. According to the invention, accurate positioning, segmentation and classification of vasculopathy can be realized without labeling data, and the automation level and accuracy of vasculopathy diagnosis are improved.

The invention relates to the field of medical image processing, in particular to a gastric cancer precancerous lesion progress risk assessment system based on time sequence image analysis, which comprises a time sequence gastroscope image registration module, a lesion curved surface dynamic representation module, a multi-dimensional feature manifold construction module, a risk dynamic prediction module and a virtual pigment endoscope enhancement module, according to the method, a differential geometry theory is introduced, the gastric mucosa surface is modeled as a Riemannian curved surface, and Gaussian curvature, average curvature and other characteristics are extracted; constructing a feature manifold space by adopting a manifold learning method, and analyzing a lesion state evolution trajectory through geodesic distance; and predicting the time for the lesion to reach a high-risk state based on the probability density distribution and the evolution vector field on the manifold. According to the method, accurate characterization of gastric cancer precancerous lesion morphological characteristics, quantitative analysis of dynamic evolution and accurate prediction of risk progress are realized, a scientific basis is provided for clinical follow-up visit decisions, and the method has the remarkable advantages of improving the early intervention rate, optimizing medical resource allocation and the like.

The invention relates to a reasoning enhanced vision-language large model training and image processing method, and the training method comprises the following steps: obtaining an ultra-wide-angle fundus image as an input image, and taking the manual annotation DR classification, the manual annotation lesion type and the clinical background of the ultra-wide-angle fundus image as cue words; utilizing a vision-language model with reasoning ability to generate reasoning enhanced image description and DR classification and lesion types obtained through reasoning; and taking the generated image description and the DR classification and lesion type obtained by reasoning as instructions, constructing a reasoning enhancement instruction data set in combination with the ultra-wide-angle fundus image, and finely adjusting the reasoning enhancement type vision-language large model. Compared with the prior art, the method has the advantages of being capable of effectively integrating clinical knowledge, high in recognition accuracy, high in interpretability and the like.

A method, system, device and storage medium for measuring and classifying CT-FFR, dividing lesions into different types according to coronary artery computed tomography angiography (CCTA) images, determining lesion lengths at lesion positions according to different lesion types, and determining the lesion lengths at the lesion positions according to different lesion types. Then, the measurement range of computed tomography fractional flow reserve is determined according to the focus lengths of different lesion types, a statistical matrix is established to calculate the numerical values of all computed tomography fractional flow reserve CT-FFR in the measurement range, and finally, hemodynamic classification is conducted on calculation results. According to the CT-FFR evaluation method, the accuracy and the stability of CT-FFR evaluation are enhanced by carrying out pathological change specificity classification on the pathological change in the haemodynamics sense, interference is reduced, and a more reliable solution is provided for clinicians and patients.

The invention relates to an intelligent target detection method and system for a gastroscope image, and the method comprises the steps: carrying out the multispectral decomposition of an obtained gastroscope image, and generating a standardized image in combination with a dynamic domain adaptation strategy; based on the standardized image, adopting a multi-scale geometric perception feature extraction network to extract fusion features; carrying out anatomical position sensing processing on the fusion features to generate position sensing features with anatomical region identification capability; on the basis of the position sensing features, morphologically-guided multi-task target detection and classification are executed, and a multi-task detection result is obtained; the multiple tasks comprise focus detection, focus boundary segmentation and focus type classification; aiming at a multi-task detection result of continuous image frames in the gastroscopy process, applying a time sequence consistency constraint, and generating a space-time optimization result; and constructing an uncertainty perception mechanism based on a space-time optimization result, and outputting a calibrated target detection result. The precision of gastroscope image detection can be improved.

The invention discloses an automatic classification method for hepatic echinococcosis based on an ultrasonic image. The method comprises the following steps: in a preprocessing stage, denoising, normalization and size standardization are carried out on an original image, and quality gating is executed; in the segmentation stage, focus positioning and pixel-level segmentation are completed on the standardized image, and a stable mask is obtained through'opening-closing 'morphology and coverage rate constraint; in the feature and initial judgment stage, multi-scale feature extraction is carried out on a focus area, and uncertainty is calculated; in the fine judgment stage, features are dynamically weighted and fused according to reliability, and focus types and activity grades are output; in the closed-loop stage, self-adaptive write-back of the model parameters, the segmentation threshold and the enhancement gain is driven by the compound loss; in the result stage, temperature scaling and expected calibration error evaluation are implemented, and a structured report containing type, classification, confidence and parameter snapshots is generated.

The invention provides a dual-order optimization self-adaptive diabetic mesh screening model generation method and lesion recognition equipment, and relates to the technical field of diabetic mesh screening, and the method comprises the following specific steps: collecting a plurality of fundus images and corresponding medical record data, carrying out fine processing and detailed labeling, presetting a machine learning detection model for training, and carrying out the recognition of the fundus images and the corresponding medical record data. The fundus image and the medical record data of the samples in the training set are used as input features, the corresponding label content is used as an output label, and dual-order optimization and self-adaptive adjustment are adopted to enhance the recognition capability of the model on lesion features; screening a model according to a category consistency coefficient between lesion types, a feature deviation coefficient between lesion features and a logic consistency coefficient of logic rules, and screening a dual-order optimization adaptive sugar mesh screening model by using samples in a test set. Clinical diagnosis can be accurately assisted, the missed diagnosis and misdiagnosis rate is effectively reduced, the problem that the feature recognition precision of a traditional model is insufficient is solved, it can be ensured that prediction conforms to clinical logic, and the reliability of the model is improved.

The application provides a blood vessel lesion determination method and device, electronic equipment and storage medium. The determination method comprises: inputting acquired no-label images and labeled images as input images into a blood vessel lesion analysis model in a single-alternating input manner; if the input image is a no-label image, training the blood vessel lesion analysis model according to a reconstruction loss function; if the input image is a labeled image, determining whether the reconstruction loss function, a lesion type loss function and a lesion degree loss function simultaneously satisfy a target condition, and if the target condition is satisfied, obtaining a trained blood vessel lesion analysis model; and determining a lesion type result and a lesion degree result of a blood vessel image according to the trained blood vessel lesion analysis model. The technical solution provided by the application can reduce manual labeling operations, reduce the workload of doctors, and ensure the accuracy of blood vessel lesion determination.

The invention provides an image detection method and device, equipment and a storage medium, and the method comprises the steps: obtaining a detection image obtained through plain-scan CT, and extracting a target body part image from the detection image; performing first image classification segmentation processing on the target body part image through a first image detection model to determine that a first target lesion type and a lesion area corresponding to the first target lesion type exist in the target body part image; and performing second image classification segmentation processing on the target body part image through a second image detection model to determine a second target lesion type and a lesion area existing in the target body part image, the second target lesion type being a sub-category of the first target lesion type. Through cooperation of the two image detection models, refined detection of an image containing a lesion type and a lesion area can be realized.

The present invention relates to a weakly supervised learning diabetic retinopathy grading and lesion identification method and system, the method comprising: S1: inputting a diabetic retinopathy fundus color image into a CNN network to obtain a feature map F; inputting F into a lesion point capture module based on weakly supervised object positioning, selecting to enter a branch module, calculating a score map, multiplying the score map with the feature map F to obtain a new feature map F'; S2: using an optimal network structure search module based on reinforcement learning to calculate the insertion position, selection probability, feature discarding threshold and retention threshold of the weakly supervised lesion point capture module in the CNN network, and finally outputting the feature map F. NA S3: Lesion attribute prediction and disease classification are obtained through attribute mining and lesion identification modules. S4: Based on the results of S3, the final lesion identification result is obtained through multiple iterative erasure calculations. The method provided by the present invention can improve the effectiveness of lesion capture and use lesion mining methods to determine lesion type, providing a basis for disease classification.

The application discloses a brain injury identification method and an identification system. The identification system identifies the injury of a patient according to the identification method. The identification method comprises the following steps: firstly, the electrical impedance tomography (EIT) of the brain of the patient is performed to obtain the EIT image of the brain of the patient; then, the region with abnormal resistivity distribution in the EIT image of the brain is taken as a reference for scanning of a near-infrared spectrometer, and a scanning path of the near-infrared spectrometer is planned; finally, the near-infrared spectrometer is scanned along the scanning path, and the lesion range and the lesion type are determined according to the obtained near-infrared spectral data. The electrodes need not be arranged on the whole head of the patient, and the automatic and rapid identification of the lesion range and the nature of the brain injury is realized.

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 provides a face acne skin lesion detection matching method, system and device, and relates to the field of image processing, and the method comprises the steps: calling a face key point detection interface, and obtaining face key point coordinates on face images at different angles; obtaining a first mapping transformation reference point of each face key point corresponding to the side face and the front face; according to the first mapping transformation reference point, generating a perspective transformation matrix of mapping the side face to the front face; according to the perspective transformation matrix and the human face key point coordinates, mapping the skin lesion detection frame on the side face to a position corresponding to the front face; matching the skin lesion detection frame on the front face with the skin lesion detection frame mapped to the front face by the side face; and unifying the skin lesion type labels of the skin lesion with the front face and the side face successfully matched. According to the invention, the accuracy of acne evaluation results can be improved.

The application discloses an auxiliary identification method for brain inflammatory pseudotumor MRI images based on machine learning, and belongs to the technical field of medical image processing, and comprises the following steps: acquiring preoperative magnetic resonance imaging (MRI) data of a brain lesion of a to-be-measured individual, and performing standardization preprocessing on the MRI data to obtain a standardized brain MRI image; a lesion target region containing a tumor core region and a peritumoral edema region is segmented on the standardized brain MRI image, and a training set, a verification set and a test set are generated based on the lesion target region; a lesion type auxiliary identification model is constructed based on feature screening and feature topology graph attention network of SVM weights, and the training set and the verification set are used for training; the test set is input into the trained lesion type auxiliary identification model to obtain a lesion type identification result, and the problems of time-consuming and labor-consuming, low efficiency, omission of hidden dangers, low accuracy and the like in existing artificial visual measurement and traditional image processing measurement methods are solved.

The invention discloses a generative artificial intelligence cerebral artery lesion detection method based on plain-scan brain CT. The method comprises the steps that 1, a plain-scan brain CT image to be processed is acquired; 2, converting the plain scanning brain CT image into a CTA image through an adaptive noise elimination network; and 3, carrying out multi-modal lesion detection on the basis of the CTA image generated in the step 2. According to the method, a CTA image is generated through an adaptive noise elimination network (ANE-NET), and a real-time lesion feature analysis engine (RTAL-FE) is embedded in the generation process, so that real-time classification prediction of lesion types is realized. And furthermore, a detection result is output through a dynamic weight decision model (DWD-M), so that the generation quality and the detection efficiency are remarkably improved. The problems that a traditional method is low in generation quality and lags behind detection are solved, and the method is particularly suitable for low-dose and non-invasive cerebrovascular disease screening scenes and has important clinical application value.

The present application relates to the technical field of medical image processing, in particular to a CT image intelligent analysis system for pneumonia auxiliary screening. The present application firstly pre-processes chest CT images and detects candidate lesion regions; then extracts topological features, deep convolution features and texture statistical features based on persistent homology theory for each candidate lesion, performs feature fusion through a multi-head self-attention mechanism to generate a unified lesion representation vector; then maps the lesion representation vector to a pre-constructed radiology knowledge graph, performs reasoning using a graph neural network, and outputs the pneumonia suspected probability and lesion classification for each lesion; finally, the evidence theory is used to fuse and quantify the uncertainty of the analysis results of multiple lesions to generate a comprehensive screening report. The present application effectively identifies complex morphological lesions through topological features, accurately identifies lesion types through knowledge graph reasoning, and provides quantitative diagnosis uncertainty through the evidence theory.

The invention relates to a method for classifying and identifying lesions in magnetic resonance images based on a causal line network model, and the method comprises the steps: collecting magnetic resonance multi-parameter sequence data, screening out sample data from the data, carrying out the preprocessing, and constructing a data set; building a causal line network model, and training the causal line network model by using the data set to obtain a focus recognition model; a current to-be-recognized magnetic resonance sequence image is preprocessed, then the image is input into the focus recognition model, and a corresponding focus type and a sign signal are output and obtained. Compared with the prior art, the accuracy and interpretability of focus type identification can be improved, and more comprehensive focus characteristic prediction, better clinical interpretability, causal interpretability of a model structure and higher prediction output consistency are realized.

This invention discloses a method and device for segmenting diabetic retinopathy based on feature interaction, comprising: constructing a feature interaction module for semantic relationship modeling of multiple lesions, embedding it into each ViT block encoding layer of the image encoder of the image segmentation model SAM, obtaining an image segmentation model FIASAM fused with the feature interaction module; training the image segmentation model FIASAM fused with the feature interaction module using fundus color image-mask annotation; acquiring fundus color images of diabetic patients, inputting them into the trained image segmentation model FIASAM fused with the feature interaction module, and generating image segmentation results containing all lesion types. This invention can enhance the feature interaction capabilities between different lesion types of diabetic retinopathy and between lesions and the background of fundus color images, thereby improving the accuracy and reliability of automatic segmentation of multiple lesions in fundus color images.

This invention relates to the field of medical data mining technology, specifically to an automatic analysis and classification system for pulmonary function test data based on big data. First, it calculates a single-channel weighted evolution index based on the variation deviation of multi-channel data time-series curves, capturing the dynamic evolution characteristics of each dimension within a continuous observation window. Then, it combines the variation trend deviation between multiple channels to accurately assess the feature response sensitivity of each dimension in response to changes in pathological states. Finally, by comparing the differences between current data and historical lesion characteristics, it quantifies the discriminative contribution of each dimension when considering historical data for each lesion type. This allows the weighted discriminative signal intensity obtained based on feature response sensitivity and discriminative contribution to more comprehensively evaluate the sample contribution of each pulmonary function test dimension. This results in higher accuracy in training the classification model using the weighted discriminative signal intensity as sample weights, thereby improving the accuracy of pulmonary function test data classification.

The application discloses an electronic medical record retrieval method and device. The method comprises the following steps: analyzing an imaging description text, determining a lesion type described by the imaging description text, and extracting an imaging sign description related to the lesion type from the imaging description text; extracting an imaging sign description related to the lesion type from the imaging description text; predicting a lesion type corresponding to a target object based on the imaging sign description; generating a retrieval prompt text based on the lesion type; retrieving a plurality of candidate medical record segments related to the lesion type based on the retrieval prompt text; and accurately retrieving medical record segments based on imaging characteristics to avoid omissions or false detections caused by ambiguous natural language expressions or incomplete retrieval conditions. Based on the importance values of the plurality of candidate medical record segments, the plurality of target medical record segments with the highest matching degree with the retrieval prompt text are selected from the plurality of candidate medical record segments, thereby improving the efficiency and accuracy of medical record retrieval.

This application provides a comprehensive non-destructive detection method for internal lesions in potatoes based on multispectral feature screening and deep learning. The method includes: S1 collecting potato samples and obtaining their spectral data; S2 selecting a combination of CARS-VIP and CARS-SPA screening adapted to absorbance spectra, and a dual-wavelength correlation coefficient method adapted to energy spectra, based on the spectral data, to obtain a subset of characteristic wavelengths; S3 constructing a three-level discrimination model based on the subset of characteristic wavelengths to identify various internal lesions in potatoes. This model sequentially distinguishes between healthy and diseased samples, further subdivides lesion types in diseased samples, and enhances discrimination through an improved residual neural network for early or boundary lesion samples. This method, through multi-strategy feature screening and a three-level discrimination model, achieves high-precision non-destructive detection of various internal lesions in potatoes, significantly improving the sensitivity of early, subtle lesion identification, and providing efficient technical support for quality control in the potato industry.

The invention discloses an intrahepatic cholangiocarcinoma or tumor conventional ultrasonic image auxiliary diagnosis method and system based on a Vision-RWKV framework, and relates to the technical field of medical image auxiliary diagnosis. Comprising the following steps: acquiring image data information which is conventional liver ultrasonic image data of a plurality of existing patients; constructing an intrahepatic cholangiocarcinoma or tumor identification model based on a Vision-RWKV architecture, wherein the model has a linear complexity bidirectional attention mechanism; training an intrahepatic cholangiocarcinoma or tumor recognition model based on a Vision-RWKV framework by adopting the image data information; and inputting a conventional liver ultrasonic image of a patient to be identified into the trained intrahepatic cholangiocarcinoma or tumor identification model to obtain an identification result, wherein the result comprises a lesion type, lesion position calibration and credibility. The method has the advantages that ultrasonic examination is simple, economical and easy to operate, and CT / MRI examination is high in accuracy, so that more and wider crowds are covered and served, missed diagnosis and misdiagnosis are reduced, and patients obtain good prognosis and life quality.

A method and apparatus for generating a lesion model of a target site, and a computer readable storage medium. The method comprises: acquiring a target lesion type of a target site; on the basis of the target lesion type of the target site, determining a plurality of target sub-anatomical structure models from a model database of the target site; on the basis of the plurality of target sub-anatomical structure models, generating a lesion model of the target site about the target lesion type; and displaying the lesion model of the target site about the target lesion type. According to the present application, a lesion model of a corresponding lesion type can be generated for the target site.

Disclosed are a method and apparatus for generating a lesion model of a tissue site, and a computer-readable storage medium. The method includes obtaining a determined lesion type of the tissue site, determining a plurality of determined sub-anatomical models from a model database of the tissue site based on the obtained determined lesion type, generating a lesion model of the tissue site based on the plurality of determined sub-anatomical models, and displaying the lesion model of the tissue site. Accordingly, the lesion model of the lesion type corresponding to the tissue site can be generated as disclosed herein.

The invention relates to the technical field of medical data mining, in particular to a lung function examination data automatic analysis and classification system based on big data, which comprises the following steps: firstly, calculating a single-channel weighted evolution index based on change deviation of a multi-channel data time sequence curve, and capturing dynamic evolution characteristics of each dimension in a continuous observation window; furthermore, the characteristic response sensitivity of each dimension in response to pathological state change is accurately evaluated in combination with data change trend deviation among multiple channels; and finally, by comparing the difference between the current data and the historical lesion features, quantizing the discrimination contribution degree of each dimension when the historical data of each lesion type is considered. The sample contribution of each lung function examination dimension can be evaluated more comprehensively by the weighted discrimination signal intensity obtained based on the feature response sensitivity and the discrimination contribution degree; the accuracy of classification model training is higher when the weighted discrimination signal intensity is used as the sample weight, and the accuracy of lung function examination data classification is improved.