Ultrasonic blood vessel intelligent analysis method and system

Abstract

The invention relates to an ultrasonic blood vessel intelligent analysis method and system. The method comprises the steps that real-time image frame stream input and preprocessing are carried out based on an ultrasonic operation request; determining whether the current image frame is an effective image frame based on the blood vessel bounding box coordinates and the confidence coefficient; the operation suggestion module inserts the current image frame into the image structure for the detected effective image frame; using a graph structure to store an image frame with effective detection; determining whether to perform a withdrawal operation; a withdrawal position and a withdrawal mode of operation are determined if necessary. According to the method, big data and an artificial intelligence technology are fused, so that personalized intelligent ultrasonic operation guidance can be carried out for different operators, module parameters are customized based on different ultrasonic operation plans, intelligent suggestion models are loaded according to stages where the different operators are located, each different module can be independently and intelligently upgraded, and the operation efficiency is improved. And the ultrasonic operation efficiency is obviously improved.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent medical technology, and in particular relates to an ultrasonic blood vessel intelligent analysis method and system. Background Art

[0002] How to use the rapidly developing big data and artificial intelligence technologies to guide ultrasound operation efficiency and customize a personalized ultrasound operation suggestion system is a technical problem to be solved. Based on the above problems, the present invention integrates big data and artificial intelligence technologies to enable personalized intelligent ultrasound operation guidance for different operators, customize module parameters based on different ultrasound operation plans, and load intelligent suggestion models according to the stage of different operators. It provides personalized ultrasound guidance paths based on the operator's operation proficiency, significantly improving the efficiency of ultrasound operation. Summary of the Invention

[0003] In order to solve the above problems in the prior art, the present invention proposes an ultrasonic vascular intelligent analysis method and system, the method comprising: Step S1: inputting and preprocessing real-time image frame stream based on ultrasound operation request; Step S2: The blood vessel quality assessment module determines whether the current image frame is a valid image frame based on the blood vessel bounding box coordinates and confidence level; The operation suggestion module determines whether the detected image frame is standard based on the clarity and morphology of the vascular imaging, inserts the current image frame into the graph structure, and gives operation suggestions; The image frame management module uses a graph structure to store valid image frames. For a valid current image frame, if it is a standard image frame, it is made a child node of all leaf nodes corresponding to the first standard image frame in its predecessor node. Otherwise, the number of predicted adjustments from the current image frame to the standard image frame is determined, and the current image frame is made a child node of the current node. The current node is updated to the node where the current image frame is located. Step S3: The image frame management module determines whether to perform a retraction operation; if so, determines a retraction position; determines a retraction operation mode based on a difference between the operation mode at the retraction position and the operation mode of the current image frame; and returns to step S1 until the planned operation time length is reached; wherein: After receiving the request to insert the graph structure, determining the path length between the node where the standard image frame closest to the current node is located and the current node, and determining to perform a retraction operation when the path length is greater than a preset path length; When the withdrawal method is maximum withdrawal, the node where the standard image frame closest to the current node is located is used as the withdrawal position; otherwise, the node with the smallest number of predicted adjustments between the node where the standard image frame closest to the current node is located and the current node is used as the withdrawal position; the node at the withdrawal position is used as the current node.

[0004] Furthermore, the medical-specific interface uses the DICOM Streaming or RTSP protocol and H.264 encoding.

[0005] Furthermore, when the graph structure is used to save the image frame, the standard score, acquisition method, blood vessel location, acquisition timestamp, etc. of the image frame are saved as node attributes.

[0006] Furthermore, the method of determining the predicted number of adjustments from the current image frame to the standard image frame is specifically as follows: analyzing the historical image frame sequence, taking the historical image frame similar to the current image frame in the historical image frame sequence as the target image frame, determining the standard image frame closest to the target image frame in the historical image frame sequence, and taking the number of historical image frames between the two as the number of adjustments; repeating the above steps until the historical image frame sequence is processed to obtain one or more adjustment times; and taking the characteristic value of the adjustment times as the predicted adjustment times.

[0007] Furthermore, the step S1 is specifically as follows: the ultrasound device pushes the image frame stream in real time through the medical-specific interface; the API gateway authenticates and formats the image frame stream, performs invalid filtering on the real-time pushed image frame stream, and pushes the filtered current image frame to the vascular quality assessment module in real time.

[0008] Furthermore, step S4: the image frame management module determines a personalized path based on the graph structure and provides it to the operator; specifically: the image frame management module traverses the graph structure from the root node, determines a personalized graph path from the root node to the leaf node, organizes the acquisition methods of all nodes involved in the personalized graph path in chronological order to obtain a personalized path; and provides the personalized path to the operator; the personalized path includes the shortest operation path, the highest quality path and / or the fastest compliance path, and the corresponding personalized graph path is the path with the shortest node span, the path with the highest standard score average and / or the shortest time path.

[0009] Furthermore, after completing the request to insert the graph structure, it is determined whether an operation retraction and / or forward operation is required.

[0010] Furthermore, the vascular quality assessment module is used to detect target blood vessels; a pre-trained vascular detection model is provided in the vascular quality assessment model; the input image frame is sent to the vascular detection model; the vascular detection model performs inference on each image frame and outputs the coordinates of the vascular bounding box and the confidence level; and based on the vascular bounding box coordinates and the confidence level, it is determined whether the current image frame is a valid image frame.

[0011] An ultrasonic blood vessel intelligent analysis system is used to implement the above-mentioned ultrasonic blood vessel intelligent analysis method.

[0012] An ultrasonic blood vessel intelligent analysis platform is used to implement the above-mentioned ultrasonic blood vessel intelligent analysis method.

[0013] The beneficial effects of the present invention include: (1) Image frame management based on a directed graph structure can obtain detection experience for the current image frame from the image frame big data containing existing monitoring experience, guide the current operator to perform standardized ultrasound section operation simulation training and form a directed graph representing a personalized operation path; and can provide a personalized path based on the directed graph to optimize ultrasound operation; (2) The integration of big data and artificial intelligence technology enables personalized intelligent ultrasound operation guidance for different operators, customizes module parameters based on different ultrasound operation plans, and loads and upgrades intelligent suggestion models according to the stages of different operators. It provides personalized ultrasound guidance paths based on the operator's operation proficiency, significantly improving the efficiency of ultrasound operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application, but do not constitute an improper limitation of the present invention. In the drawings: Figure 1 This is a schematic diagram of the ultrasonic vascular intelligent analysis method provided by the present invention.

[0015] Figure 2 A schematic diagram of a directed graph provided by an embodiment of the present invention.

[0016] Figure 3 This is a schematic diagram of the initialization of the ultrasonic vascular intelligent analysis system provided by the present invention. DETAILED DESCRIPTION

[0017] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The exemplary embodiments and descriptions are only used to explain the present invention but are not intended to limit the present invention.

[0018] The present invention proposes an ultrasonic blood vessel intelligent analysis method and system, as shown in the attached Figure 1 As shown, the method includes the following steps: Step S1: Real-time image frame stream input and preprocessing based on ultrasound operation request; specifically, the ultrasound device pushes the image frame stream in real time through the medical dedicated interface; the API gateway authenticates and formats the image frame stream, filters the pushed image frame stream for invalidity, and pushes the filtered current image frame to the vascular quality assessment module in real time; Preferably, the medical-specific interface uses DICOM Streaming or RTSP protocol and H.264 encoding; Preferably: the identity verification is a double verification based on the operator identity and / or the device ID; The real-time pushed image frame stream is invalidly filtered, specifically: dynamically extracting frames from the image frame stream; using motion-aware adaptive frame extraction; extracting 2 frames per second by default initially, and automatically increasing the frame rate to 5 frames per second when the motion amount calculated by the inter-frame difference method is greater than the preset motion amount value to determine that the probe movement is detected; and deduplicating consecutive similar frames in the extracted frames to ensure input diversity; Preferably, the image frames are organized in the form of image data files, wherein the image data files contain image data and acquisition methods, and the acquisition methods include probe posture, scanning depth, pressure parameters and / or gain, etc.; Preferably, the ultrasound operation request includes the operator identity and device ID, and an ultrasound operation plan; the ultrasound operation plan includes the planned operation time, the targeted vascular region, the scoring criteria, the preset path length, the retraction method, the acceptable retraction length, the forward feedback time, the N value, etc.; the API gateway, the vascular quality assessment module, the operation suggestion module, and the image frame management module are loaded based on the ultrasound operation request; and the ultrasound operation plan is formulated for the current operator; Step S2: The vascular quality assessment module is configured to detect target blood vessels; the vascular quality assessment model includes a pre-trained vascular detection model; an input image frame is sent to the vascular detection model; the vascular detection model performs inference on each image frame, outputting the coordinates of a vascular bounding box and a confidence score; and determining whether the current image frame is a valid image frame based on the vascular bounding box coordinates and the confidence score. Preferably: the blood vessel detection model is a YOLOv5s detection model; The determining whether the current image is a valid image frame based on the blood vessel bounding box coordinates and the confidence level is specifically as follows: determining whether a blood vessel is detected in the current image frame based on the blood vessel bounding box coordinates and the confidence level; and determining that the current image frame is valid if blood vessels are detected at least twice in the three most recent image frames including the current image frame; Preferably: when the confidence level of the output detection result is greater than 0.7, it is determined that a blood vessel is detected; Preferably: if the output confidence level is greater than 0.7, the area of the detected blood vessel region is less than 0.5% of the total image area, the blood vessel is continuously visible, and the aspect ratio is within the standard range of 1:2 to 1:4, then the blood vessel is determined to be detected; Step S2 further includes: an operation suggestion module for determining whether the detected valid image frame is standard based on the clarity and morphology of the vascular imaging, and providing an operation suggestion to the operator to assist in obtaining a high-quality image; specifically, the operation suggestion module for determining whether the detected valid image frame is standard based on the clarity and morphology of the vascular imaging, accessing the image frame management module to insert the current image frame into the image structure; and providing an operation suggestion based on feedback from the image frame management module; The method of determining whether the clarity and morphology of the vascular imaging are standard is specifically as follows: calculating the sharpness of the vascular wall edge by using the Sobel operator gradient amplitude, detecting the signal-to-noise ratio of the blood flow area by using the ratio of the signal intensity to the background noise in the ROI; determining the image clarity score based on the relationship between the edge sharpness and the sharpness threshold, and the relationship between the signal-to-noise ratio and the signal-to-noise ratio threshold; determining the vascular morphology score based on the relationship between the aspect ratio of the measured blood vessel and the ideal range, and the relationship between the smoothness of the vascular contour and the curvature change rate threshold; determining the image frame score based on the image clarity score and the vascular morphology score; determining the image frame score based on the image clarity score and the vascular morphology score; and determining the image frame score as a standard image frame when the image frame score is greater than the scoring standard; otherwise, determining the image frame as a non-standard image frame. Preferably, the scoring criteria are preset values obtained from the ultrasound operation request, which are related to the operation requirements of the current ultrasound operation request; the sharpness threshold, signal-to-noise ratio threshold, ideal range, and curvature change rate threshold are preset values; for example, the sharpness threshold is 0.8, the signal-to-noise ratio threshold is 15dB; the ideal range is 1.5~3.0; the curvature change rate threshold is Of course, the settings of these thresholds are related to the vascular area being detected and the ultrasound equipment used. Generally, these thresholds are fixed values in the operation suggestion module. The image frame management module uses a graph structure to store detected valid image frames and their acquisition methods. Specifically, upon receiving a request to insert a valid image frame into the graph structure, if the current image frame is a standard image frame, the current image frame is made a child node of all leaf nodes corresponding to the first standard image frame in its predecessor node. If the current image frame is a non-standard image frame, the predicted number of adjustments from the current image frame to the standard image frame is determined, and the current image frame is made a child node of the current node. The current node is updated to the node where the current image frame is located. The nodes in the graph structure are image frames, and the edges are the relationships between image frames. When using the graph structure to store image frames, the standard score, acquisition method, blood vessel location, acquisition timestamp, etc. of the image frame are stored as node attributes. Preferably: each node has a unique node identifier; Preferably: the root node of the graph structure is the standard image frame / image frame acquired for the first time, recording the complete acquisition method; the graph structure is a directed graph; The method of determining the predicted number of adjustments from the current image frame to the standard image frame comprises the following steps: analyzing a historical image frame sequence, selecting a historical image frame similar to the current image frame in the historical image frame sequence as a target image frame, determining a standard image frame closest to the target image frame in the historical image frame sequence, and using the number of historical image frames between the target image frame and the standard image frame as the number of adjustments; repeating the above steps until the historical image frame sequence is completely processed to obtain one or more adjustment numbers; and using a characteristic value of the adjustment number as the predicted adjustment number; since the number of historical image frames similar to the current image frame may be one or more, the obtained number of adjustments may also be one or more. Preferably: determining the similarity of the image frames is performed based on a combination of one or more of an acquisition method of the image frames, an image feature of the image frames, an image clarity score and / or a vascular morphology score; Preferably, the characteristic value of the number of adjustment times is the mean or minimum value of the one or more number of adjustment times; Alternatively: cluster the adjustment times to obtain one or more adjustment times classification sets, and use the mean of the adjustment times of the adjustment times classification set with the smallest mean as the characteristic value of the adjustment times; The historical image frame sequence is an image frame sequence consisting of valid image frames obtained when the same vascular region is tested and / or the same type of ultrasound equipment is used for testing. The historical image frame sequence can be obtained when the ultrasound equipment is tested, when a skilled operator performs the test, or when an operator with similar testing experience as the current operator performs the test. Step S3: Determine whether a retraction operation is required. If so, determine the retraction position and the retraction operation method. Specifically, the image frame management module determines whether a retraction operation is required. If so, determine the retraction position. The operation suggestion module determines the retraction operation method based on the difference between the operation method at the retraction position and the operation method of the current image frame. The retraction operation method is fed back to the operator. The process returns to step S1 until the planned operation time is reached. Preferably: after completing the request to insert the graph structure, determining whether an operation withdrawal and / or forward operation is required; The retraction operation mode is the difference between the current operation mode and the operation mode at the retraction position. The operation mode includes the probe posture and device parameters, such as the probe displacement vector, pressure adjustment amount, and device parameter changes. For example, in the retraction operation mode, the probe moves 3mm to the right, the pressure is reduced by 0.2N, and the gain is +2dB. The image frame management module determines whether a retraction operation is required, specifically by: after receiving a request to insert a graph structure, determining the path length between the node where the standard image frame closest to the current node is located and the current node, and determining that a retraction operation is required when the path length is greater than a preset path length, thereby avoiding repeated reinforcement of incorrect operation modes caused by repeated invalid operations; The determination of the retraction position is specifically as follows: when the retraction mode is maximum retraction, the node where the standard image frame closest to the current node is located is used as the retraction position; otherwise, the node with the smallest number of predicted adjustments between the node where the standard image frame closest to the current node is located and the current node is used as the retraction position; and the node at the retraction position is used as the current node; Alternatively, the node with the smallest predicted adjustment times among the nodes whose path length to the current node is less than the acceptable retraction length is used as the retraction position; further, the acceptable retraction length is less than the preset length; and a feasible operation path can be found under the operator's operating habits in the case of an erroneous operation mode; The step S3 further includes: the image frame management module determining whether a forward operation is required; if so, the operation suggestion module inputting the image frame sequence between the standard image frame closest to the current node and the current image frame into the intelligent suggestion model to obtain a suggested acquisition method; feeding the suggested acquisition method back to the operation suggestion module, and the operation suggestion module determining the difference between the suggested acquisition method and the current acquisition method to constitute a forward operation; The intelligent suggestion model is an artificial intelligence model pre-trained using a standard historical image frame sequence; the standard historical image frame sequence is an image frame sequence acquired by a skilled inspection operator when inspecting the same vascular region and / or using the same model of ultrasound equipment; The image frame management module determines whether a forward operation is required, specifically if: no standard image frame has been obtained within the forward feedback time, and N consecutive image frames are non-standard image frames. For example, in a carotid artery ultrasound examination, the vessel wall is blurred in two consecutive frames, and the most recent standard image frame is the image frame obtained 10 seconds ago. The intelligent suggestion model provides a forward operation to reduce the probe tilt angle by 2° and increase the gain by 2dB. The forward feedback time and N value are both derived from the ultrasound operation request. Alternatively: the forward feedback duration and N value are inherent settings in the image frame management module; Table 1: Example of node management in the image frame management module; Preferably, the method further comprises step S4: the image frame management module determines a personalized path based on the graph structure and provides it to the operator; specifically, the image frame management module traverses the graph structure starting from the root node, determines a personalized graph path from the root node to the leaf node, organizes the acquisition methods of all nodes involved in the personalized graph path in chronological order to obtain a personalized path; and provides the personalized path to the operator; because the operator may obtain a standard image frame through multiple paths starting from an image frame and its acquisition method when making multiple attempts, which is related to the detection methods and detection preferences of different operators; through the image frame management of the graph structure, detection experience for the current image frame can be obtained from the image frame big data containing existing monitoring experience, guiding the current operator to perform ultrasound section standardized operation simulation training; Preferably, the personalized path includes the shortest operation path, the highest quality path, and / or the fastest target-reaching path, and the corresponding personalized graph path is the path with the shortest node span, the path with the highest standard score average (based on standard score), and / or the shortest time path (based on timestamp); Preferably: the leaf node is any leaf node, some leaf nodes, or all leaf nodes; Preferably: after the planned operation time is completed, step S4 is executed; As attached Figure 2 As shown in the figure, during an operation, the standard score is set to 80 points; nodes F0 to F9 are generated, and a graph structure is constructed based on their scores; the graph structure is a directed graph; thus, the operator can form his or her own operating habits based on big data guidance while completing the ultrasound operation request; and through retracement, branch expansion and merging are formed to form new operating methods and operating habits; As can be seen, the shortest operation path is F0→F2→F4→F8, and the cumulative displacement along this path is: (+1.0-0.5+0.5,-0.3+0.4-0.1)=(+1.0x,0.0y); device parameter adjustment: gain from 65dB to 71dB, frequency stable at 7.5MHz; score change path is 85→82→88→91 points, with an average score increment of 2 points along this path; by explicitly inheriting the standard image frames, digital transmission of operation experience is achieved, and operation causality is explicitly modeled. Compared with the linear sequence management method, this supports the operator's local attempts in different ranges, improving the efficiency of ultrasound enhancement. Preferably: after completing the ultrasound operation request, an operation request completion report is generated by traversing the graph structure; the report includes: the shortest operation path, the longest operation path, the average score increment, etc.; Based on the same inventive concept, the present invention further provides an ultrasonic vascular intelligent analysis system, which is used to implement the above-mentioned ultrasonic vascular intelligent analysis method. The system includes an API gateway, a vascular quality assessment module, an operation suggestion module, and an image frame management module. Upon receiving an ultrasonic operation request, the system generates an ultrasonic task corresponding to the operator, and creates, loads, and initializes the API gateway, vascular quality assessment module, operation suggestion module, and image frame management module for the ultrasonic task. Preferably: after loading the ultrasound task, context loading is performed, and module parameter initialization settings (e.g., various thresholds and ranges in the standard scoring process) are performed based on the patient's DICOM historical data (e.g., previous vascular diameters); device preset parameters and operator habit configuration (left-handedness / right-handedness mode) are further performed based on the ultrasound operation request; Preferably: the ultrasound operation suggestion is presented synchronously with the presentation screen of the current image frame; As attached Figure 3 As shown in the figure, during the system initialization process, the ultrasound device sends an ultrasound task start request to the API gateway through the HTTP protocol. The API gateway synchronously triggers the parallel initialization of the vascular quality assessment module, the operation suggestion module, and the image frame management module. The quality assessment module loads the pre-trained vascular detection model; the operation suggestion module loads the historical image frame sequence and the intelligent suggestion model; the image frame management diagram constructs a dynamic graph structure; The quality assessment module, operation suggestion module, and image frame management module independently perform initialization tasks and synchronize progress through shared memory state. When all modules are initialized, the API gateway returns a 201 Created response containing the task ID and initial parameter configuration. A computer program (also referred to as a program, software, software application, script, or code) can be written in any form of programming language, including assembly or interpreted languages, declarative or procedural languages, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, object, or other unit suitable for use in a computing environment. A computer program may, but need not, correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program, or in multiple collaborative files (e.g., files storing one or more modules, subroutines, or code portions). A computer program can be deployed to execute on one computer or on multiple computers located at one site or distributed across multiple sites and interconnected by a communication network.

[0019] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0020] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0021] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0022] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0023] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. An ultrasonic vascular intelligent analysis method, characterized in that: The method comprises: Step S1: inputting and preprocessing real-time image frame stream based on ultrasound operation request; Step S2: The blood vessel quality assessment module determines whether the current image frame is a valid image frame based on the blood vessel bounding box coordinates and confidence level; The operation suggestion module determines whether the detected image frame is standard based on the clarity and morphology of the vascular imaging, inserts the current image frame into the graph structure, and gives operation suggestions; The image frame management module uses a graph structure to save the image frames that are valid for detection; For a valid current image frame, if it is a standard image frame, make it a child node of all leaf nodes of the node corresponding to the first standard image frame in its predecessor node; otherwise, determine the number of prediction adjustments from the current image frame to the standard image frame, and make the current image frame a child node of the current node; Update the current node to the node where the current image frame is located; Step S3: The image frame management module determines whether to perform a retraction operation; if so, determines a retraction position; and determines a retraction operation mode based on a difference between an operation mode of the retraction position and an operation mode of the current image frame; Return to step S1 until the planned operation time length; wherein: After receiving the request to insert the graph structure, determining the path length between the node where the standard image frame closest to the current node is located and the current node, and determining to perform a retraction operation when the path length is greater than a preset path length; When the withdrawal method is maximum withdrawal, the node where the standard image frame closest to the current node is located is used as the withdrawal position; otherwise, the node with the smallest number of predicted adjustments between the node where the standard image frame closest to the current node is located and the current node is used as the withdrawal position; the node at the withdrawal position is used as the current node.

2. The ultrasonic blood vessel intelligent analysis method according to claim 1, characterized in that: The medical-specific interface uses the DICOM Streaming or RTSP protocol and H.264 encoding.

3. The ultrasonic blood vessel intelligent analysis method according to claim 2, characterized in that: When the graph structure is used to save the image frame, one or more of the standard score, acquisition method, blood vessel position and / or acquisition timestamp of the image frame are saved as node attributes.

4. The ultrasonic blood vessel intelligent analysis method according to claim 3, characterized in that: The method of determining the predicted number of adjustments from the current image frame to the standard image frame comprises: analyzing a sequence of historical image frames, selecting a historical image frame similar to the current image frame as a target image frame, determining a standard image frame closest to the target image frame in the sequence of historical image frames, and using the number of historical image frames between the target image frame and the standard image frame as the number of adjustments; Repeat the above steps until the historical image frame sequence is processed to obtain one or more adjustment times; and use the characteristic value of the adjustment times as the predicted adjustment times.

5. The ultrasonic blood vessel intelligent analysis method according to claim 4, characterized in that: The step S1 is specifically as follows: the ultrasound device pushes the image frame stream in real time through the medical-specific interface; the API gateway authenticates and formats the image frame stream, performs invalid filtering on the real-time pushed image frame stream, and pushes the filtered current image frame to the vascular quality assessment module in real time.

6. The ultrasonic blood vessel intelligent analysis method according to claim 5, characterized in that: Step S4: The image frame management module determines a personalized path based on the graph structure and provides it to the operator; specifically: the image frame management module traverses the graph structure from the root node, determines a personalized graph path from the root node to the leaf node, organizes the acquisition methods of all nodes involved in the personalized graph path in chronological order to obtain a personalized path; and provides the personalized path to the operator; the personalized path includes the shortest operation path, the highest quality path and / or the fastest compliance path, and the corresponding personalized graph path is the path with the shortest node span, the path with the highest standard score average and / or the shortest time path.

7. The ultrasonic blood vessel intelligent analysis method according to claim 6, characterized in that: After completing the request to insert the graph structure, determine whether an operation needs to be withdrawn and / or advanced.

8. The ultrasonic blood vessel intelligent analysis method according to claim 7, characterized in that: The vascular quality assessment module is used to detect target blood vessels. The vascular quality assessment model is provided with a pre-trained vascular detection model. The input image frame is sent to the vascular detection model. The vascular detection model performs inference on each image frame and outputs the coordinates of the vascular bounding box and the confidence level. Based on the vascular bounding box coordinates and the confidence level, it is determined whether the current image frame is a valid image frame.

9. An ultrasonic vascular intelligent analysis system, characterized in that: The ultrasonic blood vessel intelligent analysis system is used to implement the ultrasonic blood vessel intelligent analysis method according to any one of claims 1 to 8.

10. An ultrasonic vascular intelligent analysis platform, characterized in that: The ultrasonic vascular intelligent analysis platform is used to implement the ultrasonic vascular intelligent analysis method according to any one of claims 1 to 8.

Images