197 results about "DICOM" patented technology

The invention provides an index construction method for a medical ultrasonic database, and relates to the technical field of ultrasonic data processing, and the method comprises the steps: reading a multi-frame image sequence from a DICOM ultrasonic database, and generating a diagnosis intention vector for each frame; analyzing the diagnosis report, and segmenting the report into a plurality of segments; lesion detection and segmentation are performed on each frame of image to obtain lesion information, and a cross-frame aggregation strategy is adopted to identify a unified lesion object; extracting a feature vector of each focus object to obtain a focus object feature vector; establishing an alignment mapping between the focus object and the report fragment by adopting triple constraints, and generating a focus report mapping table; and constructing a multi-layer index structure, and storing the focus report mapping table and the multi-layer index structure in parallel. According to the method, a multi-layer index structure comprising a metadata index, a text index, a focus object vector index and a cross-modal embedding index is formed, so that efficient organization and focus-level accurate retrieval of medical ultrasonic data are realized.

The invention relates to the technical field of intelligent recommendation, in particular to an intelligent inquiry recommendation method and system, and the method comprises the following steps: extracting text organ word frequency features through TF-IDF, generating a matching degree matrix through cosine similarity, screening core features, analyzing positioning parameters through DICOM, constructing a co-occurrence matrix through collaborative filtering, screening associated feature pairs, indexing a recommendation library, and predicting a symptom path through LSTM. And generating a time sequence weight feature set and a PageRank iterative sorting recommendation table. According to the method, a multi-modal association system is constructed by fusing text features and image parameters, semantic tags and space coordinates are combined to filter and screen organ domain features in a collaborative manner, the matching precision of symptoms and resources is improved, a time sequence weight reconstruction model is adopted to capture symptom evolution features, and hidden state transition is adopted to enhance the disease course prediction capability; a three-dimensional decision model is constructed through feature node sorting, time, space and feature dimension unification is achieved, a personalized diagnosis and treatment scheme is optimized by fusing multi-dimensional features and dynamic weights, and the credibility and clinical applicability of a recommendation result are improved.

The invention provides a DICOM (Digital Imaging and Communications in Medicine) file sequence processing method, a medical image analysis method, system and device, a computer storage medium and a computer program product, the processing method comprises the following steps: loading a DICOM file sequence corresponding to a series of image sets obtained by tomography, and arranging according to a scanning sequence to obtain an initial image file queue; reading DICOM files from a specified position to form a key image file queue with a fixed length; and reading sequence-level DICOM metadata and image-level DICOM metadata for representing key features from the key image file queue, and arranging the sequence-level DICOM metadata and the image-level DICOM metadata according to a preset rule to obtain a sequence DICOM metadata queue with fixed length and fixed information items of each position. By adopting the scheme provided by the invention, the obtained metadata queue with the fixed dimension provides a data basis for subsequent deep learning network input, and improvement of the adaptability with the deep learning network and improvement of the analysis accuracy are facilitated.

The invention relates to DICOM image processing, in particular to a DICOM image preprocessing and quality control system based on Spark, and the system comprises a data collection module which collects a DICOM medical image from a data source, and automatically recognizes and extracts a key metadata field in the DICOM medical image; the index construction module is used for constructing a comprehensive and accurate index system according to key metadata fields of the DICOM medical images; the data standardization module is used for carrying out standardization processing on key metadata fields of the DICOM medical images, supporting version control and audit tracking on the key metadata fields and carrying out optimization processing on the DICOM medical images; the scoring module is used for carrying out objective quality evaluation on the DICOM medical image based on a multi-dimensional image quality evaluation model and calculating a corresponding comprehensive quality score; according to the invention, the defect that the DICOM medical images from different equipment manufacturers are difficult to efficiently preprocess can be overcome.

The invention provides a medical platform management system, and relates to the technical field of medical health. Comprising a multi-source heterogeneous data intelligent acquisition module, the multi-source heterogeneous data intelligent acquisition module is connected with an external system through a distributed data acquisition framework and collects and integrates multi-source heterogeneous data, and the multi-source heterogeneous data comprises but is not limited to medical text data, multi-modal medical image data and real-time medical sensor data; aiming at the problem, the multi-source heterogeneous data intelligent acquisition module designs an intelligent data interface adapter, supports multiple medical data transmission protocols such as RESTAPI, HL7FHIR, DICOM and the like, realizes data standardized mapping through an efficient data conversion engine, solves the problem of cross-mechanism and cross-system data intercommunication in a medical body, and improves the data transmission efficiency. And a foundation is laid for realizing real medical integration.

The invention discloses an intelligent endoscope system with a multi-mode imaging function, and relates to the technical field of medical equipment, and the intelligent endoscope system comprises an imaging module which comprises a white light imaging subunit, a fluorescence imaging subunit and an infrared thermal imaging subunit; the data preprocessing module comprises noise suppression, SIFT registration and normalization; the image fusion module is used for decomposing four layers in a Laplacian pyramid, and reconstructing after feature weighting; the intelligent analysis module is used for extracting multiple types of features, performing CNN-Transform model reasoning and performing historical case verification; the display control module is used for 4K display, AR labeling and touch operation; and the data storage module is used for AES-256 encryption, DICOM conversion and three-level permission. The diagnosis accuracy is improved through multi-modal fusion, lesions are analyzed and efficiently recognized, operation risks are reduced through operation navigation, safety is guaranteed through data double backup, the system is suitable for multi-cavity diagnosis and treatment, and clinical precise diagnosis and safe operation are assisted.

The invention discloses a medical image optimization method and system based on vascular branch selective blurring. The method comprises the following steps: carrying out preprocessing and blood vessel enhancement on a three-dimensional angiography DICOM image; performing precise blood vessel segmentation by using an encoder-decoder network comprising a wide activation and residual cavity space pyramid module; identifying an interference branch which shields the target blood vessel structure at a specific working angle through blood vessel topology analysis; selective blurring repair is performed on the interference branches based on a Poisson equation and a bidirectional convolution LSTM to generate an optimized image which is visually unobstructed and keeps topological continuity. And a safety mechanism of virtual-real combined display and multi-angle plan planning is introduced, so that the reliability of surgical navigation is ensured. And finally, integrating the optimized 3D blood vessel model to a radiography system supporting real-time synchronization, and using the 3D blood vessel model as a road map. The clinical problem that the optimal working angle cannot be used due to blood vessel shielding can be effectively solved, and the precision and safety of an endovascular interventional operation are remarkably improved.

The invention relates to the technical field of medical image processing, in particular to a DICOM sequence intelligent normalization and sorting method and system, and the method comprises the steps: obtaining an original DICOM file; analyzing a key DICOM tag of the original DICOM file to obtain a direction tag and body position information, and based on the direction tag, realizing normalization based on a patient coordinate system and performing direction correction; projecting the coordinates of all the slices to a normal vector direction to reconstruct a layer sequence, and counting a slice interlayer spacing sequence to estimate a real interlayer spacing based on adjacent projection difference values; identifying and rejecting abnormal slices based on the abnormal comparison target label and the slice interlayer spacing sequence; and for a multi-sequence set of the same check, extracting a sorting feature vector corresponding to each sequence, and selecting a sorting rule in combination with scene complexity to obtain a final sorting key for representing a sorting result of the phase dimension and / or the time dimension. The method has the effect of solving the problems that existing DICOM sequences are inconsistent in direction, disordered in layer sequence, difficult in time phase recognition and the like.

The invention discloses a medical image intelligent diagnosis and three-dimensional visualization interaction method and system based on artificial intelligence and Unreal Engine. The system is composed of a data acquisition module, an artificial intelligence analysis module, a three-dimensional reconstruction module, a rendering interaction module and a report generation module. The method comprises the following steps: firstly, carrying out voxel normalization on a DICOM / NIfTI image; the improved 3D U-Net is combined with CBAM attention and WGAN-GP training, and a focus initial density field is output; an AI enhanced implicit voxel reconstruction (AEIVR) algorithm is proposed, an enhanced density field is generated by using position coding + MLP, and grid extraction is completed by using improved Marking Cubes of a dynamic threshold; high-fidelity real-time rendering is realized in combination with a Nand micro polygon and Lumen global illumination in Unreal Engine, and VR / AR interaction and virtual surgery (real-time updating of resection mask driving) are supported. According to the method, the problems of large artifacts and insufficient interactivity and real-time performance of existing two-dimensional output and three-dimensional reconstruction are solved, and the end-to-end segmentation-reconstruction-rendering-interaction-reporting process is realized. On typical CT / MRI data, the segmentation Dice is about 0.90-0.94, the reconstruction voxel resolution is 0.1-0.5 mm, the artifacts are reduced by about 20%-30%, the PSNR is improved by about 3-7 dB, and the rendering interaction delay lt is achieved; the system is suitable for scenes such as lung CT, brain MRI, orthopaedic X-ray and cardiac ultrasound, and can be expanded to preoperative planning, remote consultation and teaching.

The invention discloses an intelligent patient safety risk reminding method and system based on PACS and RIS, and the method comprises the steps: collecting the multi-source medical data of HIS, EMR, LIS and the like in real time through an HL7 and DICOM multi-protocol fusion technology, constructing a dynamic evaluation model covering the core risks of allergy, metal implants, infectious diseases and the like, and carrying out the native integration of a hierarchical reminding mechanism at PACS and RIS workstations. The system adopts a Bayesian network model to realize dynamic adjustment of risk levels, a differentiated reminding interface is embedded through a webpage style technology, and risk reminding of each level forcibly requires a doctor to confirm and process. And the closed-loop feedback module records the medical care operation and returns the medical care operation to a related system, and continuously optimizes the risk assessment model. The problems that an existing system is isolated in data, single in reminding function and incomplete in risk coverage are solved, the clinical risk management level is remarkably improved, and the medical accident rate is reduced.

The invention relates to a spine bone structure and bone mineral density automatic measurement method based on a CT image. The method comprises the following steps: performing CT scanning on a phantom with known bone mineral density to obtain a real bone mineral density calculation formula; all CT images in the DICOM format are read, DICOM analysis is carried out, and three-dimensional CT value data and other main information of all the images are obtained; performing outer contour segmentation on the whole spine in the CT image by using a deep learning network to obtain an outer contour of each vertebral body in the image; the method comprises the following steps: carrying out preliminary segmentation on cortical bones and cancellous bones by adopting a dynamic threshold method aiming at each vertebral body, selecting a preliminarily segmented cancellous bone region, using a maximum connected domain algorithm in a three-dimensional direction, retaining a maximum connected region, and further measuring the bone mineral density, namely calculating overall topology parameters, the bone mineral density and the bone mineral content of the current vertebral body; calculating the rubbing parameters, the bone mineral density and the bone mineral content of the current vertebral cortical bone and cancellous bone; and measuring the content of the cross section of the cone.

The invention relates to a medical image retrieval method and system based on a DICOM protocol and a storage medium, by setting a unified proxy forwarding service and combining with an image prefetching and caching mechanism linked with a clinical business event, the configuration difference of a back-end heterogeneous PACS system is shielded, single access point connection of a client is realized, and the user experience is improved. The complexity of system deployment and maintenance is reduced; meanwhile, the image data are pre-stored in the cache, so that the retrieval waiting time of the remote image is shortened, and the data access performance and the user experience are improved; and centralized authority authentication and log auditing are realized through the agent layer, so that the controllability and the security of a cross-mechanism data sharing process are guaranteed under the condition of avoiding high cost of centralized storage of full data.

The system brings together patient data both clinical records and imaging studies from disparate sources to the user workstation or mobile device in real-time and on-demand. In order to do so, the system needs to establish application layer connectivity utilizing HL7 or FHIR and DICOM for imaging. Once a secure connection is established, the system is able to search and retrieve records and present it to end user.

The invention discloses a medical image data treatment method and system, and relates to the field of image processing, and the method comprises the steps: firstly obtaining an original DICOM file stream, carrying out the deep analysis of the original DICOM file stream through a DICOM analysis library, and extracting an analyzed DICOM object; then, a key rapid path distribution mechanism is introduced, and based on a preset meta-rule set, whether each analyzed DICOM object should enter a rule engine path or an ML path is intelligently judged; for a rule engine path, realizing standardization of a DICOM tag through a matching rule; for an ML path, semantic feature extraction and context interaction are performed on a DICOM tag by using an NLP technology, and intelligent classification with confidence is realized through a hybrid classification model. According to the double-path parallel processing, the optimal processing mode can be selected according to the data characteristics, and the data governance efficiency, the standardization degree and the intelligent recognition capability are remarkably improved.

The invention provides a target focus identification method, system and device based on a medical image, a computer storage medium and a computer program product, and the method comprises the steps: generating an MPR feature vector and a DICOM metadata feature vector which are aligned in dimension based on medical image coding, the MPR feature vector comprises the form and spatial position information of an anatomical structure, and the DICOM metadata feature vector comprises the form and spatial position information of the anatomical structure; the DICOM metadata feature vector comprises scanning parameters, equipment information and patient body position information; the MPR feature vector and the DICOM metadata feature vector are fused through a bidirectional cross attention mechanism, and a comprehensive feature vector containing lesion semantics and space details is generated; and processing the comprehensive feature vector by using a neural network model based on a Transform architecture, and outputting an identification result of the target focus. By adopting the scheme provided by the invention, the complementation of semantic information such as anatomical morphological characteristics and scanning parameters is realized, and the recognition precision of the target focus is improved.

The invention discloses a lossless switching and synchronizing system for multi-modal operation image signals, which relates to the technical field of multi-modal operation image signal processing and comprises an optical flow field coding module, a scene perception decision engine, a time-sharing imaging control module, a data co-processing interface and an exception handling unit. The optical flow field coding module constructs a unified space-time grid, and maps ultrasonic, endoscope and MRI images to grid nodes through cross-modal optical flow field calculation to generate a synchronous frame sequence; the scene perception decision engine dynamically outputs an optimal image mode combination instruction based on instrument position and tissue deformation strain rate data; the time-sharing imaging control module analyzes the instruction and generates imaging equipment excitation signal time sequence configuration; and the data co-processing interface encapsulates the synchronization frame sequence into a DICOM-RT format and transmits the synchronization frame sequence to the display terminal. The abnormality processing unit monitors signal abnormality and generates a compensation frame. According to the system, high-precision space-time synchronization and dynamic intelligent switching of multi-modal images are realized, and the precision and reliability of surgical navigation are improved.

The invention discloses an OpenMC-based BNCT dose full-process automatic calculation output method, and particularly relates to the technical field of boron neutron capture therapy dose calculation, and the method comprises the steps: reading a DICOM-CT image, automatically completing HU value segmentation, material assignment and voxel grid modeling, and generating an OpenMC geometry and material input card; setting a neutron / photon source, an operation mode and a counting grid, and executing Monte Carlo transport calculation; an ICRP flux-dose conversion coefficient library is built in, and dose calculation is automatically completed; the dose field is written into an Rt dose file, and the Rt dose file can be directly imported into mature medical image processing software for visual evaluation and plan optimization. According to the method, end-to-end automation from the CT image to the clinical available dose file is achieved, manual intervention is not needed, the BNCT dose calculation efficiency, precision and standardization level are remarkably improved, and a reliable tool is provided for BNCT dose simulation research.

The application provides a lung CT image processing method and system based on multi-scale space semantic fusion, which comprises the following steps: collecting multiple lung CT-DICOM images, and performing quality inspection processing and format conversion to obtain original sample images; pre-processing and lung parenchyma segmentation are performed on the original sample images to obtain standard sample images; positive and negative samples are determined according to the image labeling results, and the positive and negative samples are subjected to random interference to obtain expanded sample images; a training data set is generated based on the expanded sample images and the standard sample images; the YOLOv12 network is improved based on the MSSF module and the C3K2_MSSF module to obtain an improved YOLOv12-3D network, and the improved YOLOv12-3D network is trained by using the training data set to obtain a target detection model; and the CT image to be detected is input into the target detection model to obtain detection information. The improved YOLOv12-3D network architecture realizes deep fusion of three-dimensional space features and multi-scale space semantic features in the feature extraction stage, and significantly improves the extraction rate and efficiency of small targets.

The present application relates to the technical field of image segmentation, and provides a lung image segmentation method, which comprises the following steps: S1: inputting a lung CT dicom sequence and converting the original dicom sequence into a jpg format image as an original lung image; S2: enhancing the original lung image through an image enhancement Lap-CLAHE algorithm to obtain an enhanced image; and S3: inputting the enhanced image into a U-shaped symmetrical network segmentation model 3D-SDUnet composed of an encoder, a decoder, a DPA double-path attention and a skip connection, and outputting a final segmentation result of the lung image. According to the technical scheme, the lung image is first enhanced, the detail information is improved, and the good image contrast is maintained, so that high-quality input data is provided for subsequent model segmentation. The segmentation model is optimized, the skip connection composed of a dense Swin Transformer block (DATB) and the DPA attention module are added, the information connection between the encoder and the decoder is strengthened, the segmentation precision is improved, and a complete lung model is obtained.

The application provides an image segmentation method and device, the image segmentation method comprises the following steps: obtaining a DICOM image to be segmented; performing feature extraction on the DICOM image based on a multi-layer perception machine, obtaining a plurality of sampling feature maps, and performing convolution operation on a target sampling feature map to obtain a multi-scale feature map; and obtaining a segmentation image based on the sampling feature map with the highest image resolution and the multi-scale feature map. The method can obtain the width information and height information of the image by obtaining a plurality of sampling feature maps of the DICOM image through the multi-layer perception machine, and can obtain the abstract information of the image by performing convolution operation on the target sampling feature map to obtain the multi-scale feature map. Finally, the sampling feature map with the highest image resolution and the multi-scale feature map are fused to realize the splicing of the feature information of different dimensions of the image, and the segmentation image is obtained according to the fused feature map, thereby improving the image segmentation accuracy.

Provided are an apparatus and method for verifying forgery of a medical image using a hash value based on meta information of the medical image.SOLUTION: The forgery verification apparatus 400 includes a memory configured to store patient-specific DICOM files and a processor functionally connected to the memory, and the processor is configured to, when issuance of a DICOM file is requested, search for the requested DICOM file among the patient-specific DICOM files, generate an issuance number associated with the issuance, extract a part of meta information of the DICOM file, generate a first hash value using the part of the meta information and the issuance number, and store the first hash value in the memory in association with the issuance number. While providing a copy of a DICOM file to an external storage device, a first hash value is inserted into a designated tag area of the copy as forgery verification information.SELECTED DRAWING: Figure 1

Disclosed herein are methods for selecting and preparing patient data to facilitate the adoption and customized training of machine-learning models in clinical settings, particularly for radiation therapy treatment planning. The disclosed embodiments streamline the customized training process through an automated workflow that includes prefiltering patient metadata, retrieving relevant DICOM files, optional data anonymization, and generation of training data. The data is then organized into a format suitable for machine-learning training. The embodiments discussed herein reduce manual labor, minimize errors, and accelerate the integration of machine-learning into clinical workflows, enabling clinics to train and implement predictive models that replicate specific clinical practices, thereby enhancing treatment precision and improving patient outcomes.

The invention discloses a display characteristic curve correction method and system for a medical LED display screen. The method comprises the first step of full-gray-scale range brightness data collection, the second step of pixel-level response curve fitting, the third step of medical standard parameter calculation, the fourth step of real-time dynamic correction application and the fifth step of intelligent periodic maintenance management. According to the invention, through point-by-point correction and curve fitting in the whole gray scale range, high-precision display characteristic compensation is realized, and the uniformity and stability of the display screen are ensured; medical standards such as DICOM, GSDF and the like are strictly followed, and the reliability and diagnosis accuracy of medical image display are improved; the automatic periodic correction mechanism reduces manual intervention, improves the maintenance efficiency and prolongs the service life of the display screen.

The present application is suitable for the technical field of medical signal processing, and provides a shape-driven two-dimensional Cine-MRI small intestine movement evaluation method. The present application shows significant effects in the aspects of small intestine movement quantitative evaluation and clinical application. The present application takes the geometric continuity of the intestinal cavity as a constraint, automatically segments, space-time filters, and reconstructs a closed and hole-free space-time shell through a weighted level set, extracts a time-consistent center line and optimizes the processing, and then converts the cavity wall movement through radiation encoding and constructs four types of quantitative indexes. The method can distinguish between healthy and disease groups, the parameter difference is significant and the consistency with expert annotation is high, can provide a reviewable quantitative baseline for efficacy follow-up, and can also complement the MRE index to improve the stability of small intestine comprehensive evaluation. Without changing the acquisition protocol, the device sensitivity is low, the single GPU near real-time processing is easy to integrate, a multi-module system can be constructed to adapt to the DICOM process, and the reliability of the results can be ensured by distinguishing real deformation from interference factors.

This application discloses a method and related products for training a generative model for generating three-dimensional aneurysms. The method includes: extracting at least one perspective of two-dimensional image data from original three-dimensional image data; semi-transparentizing the aneurysm-bearing artery region in the two-dimensional image data and marking the centerline of the aneurysm-bearing artery to obtain preprocessed data; extracting aneurysm-related two-dimensional and three-dimensional features using a pre-trained image encoder and shape encoder based on the preprocessed data and the original three-dimensional image data; and training the generative model using the two-dimensional and three-dimensional features as the input and output, respectively, so that the generative model outputs a three-dimensional aneurysm containing the aneurysm-bearing artery region and surrounding vessels. Using the solution of this application, high-precision three-dimensional aneurysm reconstruction can be achieved without original DICOM data, providing a reliable aneurysm-assisted diagnosis and treatment decision-making solution.

The method of obtaining DICOM files may comprise receiving, by a processor, scanning data of a scannable graphic associated with an item, wherein the scannable graphic is associated with DICOM tag data; extracting, by the processor, the DICOM tag data from the scanning data; determining, by the processor, that the DICOM tag data is associated with a first DICOM file; identifying, by the processor, a user of the first DICOM file; retrieving, by the processor, the first DICOM file; searching, by the processor, for other DICOM files that are associated with at least one of the user or the first DICOM file; retrieving, by the processor, the other DICOM files; and sending, by the processor and to the user, the first DICOM file and the other DICOM files.

The invention belongs to the technical field of orthopedic traction, and discloses an intelligent orthopedic traction method, system and device. The method comprises the steps that clinical data and DICOM images are collected, anatomical parameters are extracted through three-dimensional reconstruction, and an initial scheme set is obtained through a parameter recommendation model in combination with the clinical data; original signals are drawn, collected and preprocessed through the initial parameters to obtain biomechanical characteristics, and the biomechanical characteristics are input into a fear analysis model to obtain the fear level; counting and analyzing the original signal after traction to obtain a muscle fatigue result; evaluating a curative effect based on an original signal and a daily fatigue result, optimizing the digital twins by combining biomechanical characteristics and an evaluation result, and obtaining an optimized traction scheme by combining a reinforcement learning optimization scheme; according to the invention, collaborative optimization of orthopedic traction treatment on physiological recovery and psychological care is realized, and the personalized treatment level and rehabilitation effect are remarkably improved.

The invention discloses a three-dimensional surgical risk model reconstruction method based on brain anatomy and a functional atlas, and the method comprises the steps: S1, carrying out the image processing and fine segmentation of a brain MRI image or a DICOM sequence image of a patient, and recognizing a plurality of anatomical structures; s2, establishing a multi-dimensional risk assessment model, and performing quantitative scoring on each brain region from four dimensions of functional importance, structural vulnerability, recovery potential and clinical significance; s3, constructing a continuous three-dimensional risk field based on the risk score, and realizing the propagation of the risk value in the brain tissue based on a Gaussian diffusion model; and S4, based on an improved path planning algorithm, searching an optimal operation path in the three-dimensional risk field to realize risk minimization. The invention further discloses a corresponding system, electronic equipment and a computer readable storage medium.

The application discloses a kind of based on the safe processing and transmission method of DICOM file, client and server, comprising: obtaining DICOM file, the DICOM file includes metadata info and image data img;The metadata info is de-sensitized and is handled, and sensitive data is obtained;The sensitive data is encrypted, and the sensitive data after encryption is written in image data img;The metadata info after de-sensitization processing and the image data img of writing encrypted sensitive data are generated new DICOM file according to preset format, and the new DICOM file is sent to server.The application retains the IMG data and META data of original file, and is stored in somewhere of file itself by safe encryption means, and processed DICOM cannot view METADATA and other private information by ordinary method, but using corresponding authorized reverse method can still analyze and view relevant information.Therefore, the method retains META information under encryption mode, while patient privacy is maximally protected.

The application provides a kind of joint image segmentation method, device, electronic equipment and storage medium, it is related to medical image processing field, the method comprises: obtaining the DICOM image of knee joint;DICOM image is input into knee joint image segmentation network, obtains the knee joint segmentation image of knee joint image segmentation network output;Wherein, knee joint image segmentation network includes image segmentation model, edge segmentation network and image fusion module;Image segmentation model is used to carry out knee joint segmentation processing to DICOM image, obtains first segmentation image;Edge segmentation network is used to carry out edge prediction to first segmentation image, obtains second segmentation image;Image fusion module is used to fuse first segmentation image and second segmentation image, obtains knee joint segmentation image.The joint image segmentation method, device, electronic equipment and storage medium provided in the application can improve the accuracy of knee joint image segmentation.