A varicose vein data processing method, electronic device and product

Abstract

The application discloses a varicose vein data processing method, an electronic device and a product. The method comprises the following steps: acquiring a real-time ultrasonic image of a target part of a to-be-tested object, determining a vein key point in the target part based on the real-time ultrasonic image, and acquiring reference information of the vein key point. The reference information comprises a type of the vein key point, a related vein segment, a related limb segment and position information. The type of the vein key point comprises a vein reflux point and a vein backflow point. The method further comprises the following steps: identifying a vein shunt of the target part based on the reference information of the vein key point, determining a shunt type of the vein shunt, and determining and prompting a ligation point of the target part. The vein shunt is an abnormal blood flow that causes varicose veins. The application can assist in realizing the whole process from scanning to decision-making in preoperative evaluation of varicose veins, reduces the dependence on the professional skills of operators, and improves the preoperative examination efficiency of varicose veins.

Description

Technical Field

[0001] This application relates to the field of medical image processing technology, and in particular to a method, electronic device and product for processing varicose vein data. Background Technology

[0002] Varicose veins of the lower extremities are a common venous return disorder in vascular surgery. In the current treatment options, targeted ligation to block the pathological venous reverse flow (i.e., venous reflux) is usually regarded as a safer treatment option for varicose veins of the lower extremities.

[0003] Since targeted ligation surgery is performed on the patient's lower limb surface, it is necessary to accurately correspond the intravascular pathological sites found by ultrasound scan with the patient's body surface location. Therefore, during the simultaneous ultrasound scan, surgeons usually use a waterproof pen to directly mark the corresponding anatomical sites on the patient's lower limb skin surface to provide accurate surface positioning guidance for the targeted ligation operation during surgery.

[0004] However, targeted ligation places extremely high demands on surgeons' ability to dynamically interpret ultrasound images, logically deduce hemodynamic pathways, and map spatial anatomical structures in real time, which seriously hinders the clinical application and popularization of this technology. Summary of the Invention

[0005] In view of the above-mentioned defects or deficiencies in the prior art, it is desirable to provide a method, electronic device and product for processing varicose vein data, which can identify key vein points in real-time ultrasound images, determine the shunt type of venous shunt based on the reference information of the key vein points, and determine the ligation point based on the shunt type of venous shunt. This assists in realizing the entire process from scanning to decision-making in the preoperative assessment of varicose veins, shortens the training time for operators, reduces the dependence on the professional skills of operators, and improves the efficiency of preoperative examination of varicose veins.

[0006] In a first aspect, this application provides a method for processing varicose vein data, the method comprising: Real-time ultrasound images of the target area of ​​the subject are acquired, and key venous points in the target area are determined based on the real-time ultrasound images. Reference information of the key venous points is obtained. The reference information includes the type of key venous point, related venous segment, related limb segment, and location information. The types of key venous points include venous reflux points and venous return points. Based on reference information of key venous points, identify venous shunts in the target area and determine the shunt type, which is the abnormal blood flow that causes varicose veins; The ligation point for the target site is determined and indicated based on the shunt type.

[0007] In conjunction with the first aspect, in one possible implementation, key vein points in the target area are determined based on real-time ultrasound imaging, and reference information of these key vein points is obtained, including: In real-time ultrasound images, shunt and reflux indicators are annotated; shunt indicators are used to characterize abnormal venous proximity, and reflux indicators are used to characterize venous reflux. The system receives a first confirmation operation from the operator based on the shunt and reflux indications; the first confirmation operation is used to mark the venous reflux point and venous return point on the real-time ultrasound image. Obtain the type, related vein segment, related limb segment, and location information of key vein points.

[0008] In conjunction with the first aspect, in one possible implementation, obtaining the location information of key vein points includes: The venous reflux points and venous return points marked on real-time ultrasound images are mapped to real-time optical images of the target site to obtain the location information of the venous reflux points and venous return points.

[0009] In conjunction with the first aspect, one possible implementation involves annotating shunt indications on real-time ultrasound images, specifically including: The distribution of veins in real-time ultrasound images is identified. When the spatial distance between any two veins is less than a preset threshold, a shunt indicator is marked at a preset position in the real-time ultrasound image.

[0010] In conjunction with the first aspect, one possible implementation involves identifying the distribution of veins in real-time ultrasound images, specifically as follows: The distribution information of each vein in the real-time ultrasound image is obtained. The distribution information of each vein is obtained by the operator through annotation or determined based on a pre-trained first model. The input of the first model is the real-time ultrasound image, and the output is the distribution information of each vein in the real-time ultrasound image.

[0011] In conjunction with the first aspect, in one possible implementation, the reflux indication includes multiple reflux indications with different warning levels. Specifically, the reflux indication is annotated on real-time ultrasound images as follows: Multiple venous refluxes were identified in real-time ultrasound images, and the reflux time and reflux area of ​​each venous reflux were obtained; For each venous reflux, the warning level of venous reflux is determined based on the reflux time, and the reflux indication corresponding to the warning level is marked in the reflux area of ​​venous reflux.

[0012] In conjunction with the first aspect, one possible implementation involves identifying multiple venous refluxes in real-time ultrasound imaging, specifically: Identify the reflux region of each venous reflux in real-time ultrasound images; Select the points to be analyzed in each reflux region of venous reflux; Pulse Doppler waveform analysis was performed on each point to be analyzed to obtain the reflux time of each venous reflux.

[0013] In conjunction with the first aspect, one possible implementation involves obtaining the type of vein key point, the relevant vein segment, the relevant limb segment, and location information, specifically: The first prompt message is displayed to prompt the operator to mark the relevant venous segment, relevant limb segment, and location information of the venous reflux point in the real-time optical image, as well as the relevant venous segment, relevant limb segment, and location information of the venous return point in the real-time optical image.

[0014] In conjunction with the first aspect, one possible implementation involves mapping the venous reflux points and venous return points marked on real-time ultrasound images to the real-time optical images of the target site, specifically: Upon receiving the first confirmation operation, acquire the current position information of the ultrasonic probe at the target site; Based on the current location information, venous reflux points and venous return points are mapped onto real-time optical images.

[0015] In conjunction with the first aspect, in one possible implementation, after mapping the venous reflux point and venous return point to the real-time optical image based on the current location information, the method further includes: The system receives dragging operations from the operator on the currently marked venous reflux point or venous return point on the real-time optical image. The dragging operation is used to mark the venous reflux point or venous return point at the position corrected by the operator.

[0016] In conjunction with the first aspect, one possible implementation involves mapping the venous reflux points and venous return points marked on the real-time ultrasound image to the real-time optical image of the target site, specifically: A second prompt message is displayed to guide the operator to drag and map the venous reflux points and venous return points marked on the real-time ultrasound image to the real-time optical image of the target site.

[0017] In conjunction with the first aspect, one possible implementation method also includes: In response to the first confirmation operation, real-time ultrasound images and the operator's real-time scanning operation at the target site are recorded; Create markers for reference information, recorded real-time ultrasound images, and real-time scanning operations in real-time optical imaging.

[0018] In conjunction with the first aspect, in one possible implementation, the shunt type of venous shunt is determined based on reference information of venous key points, specifically as follows: Determine the blood flow path of venous shunts based on reference information from key venous points; The classification rules are obtained, and the shunt type of venous shunt is determined based on the classification rules. The classification rules include multiple shunt types and the blood flow path corresponding to each shunt type.

[0019] In conjunction with the first aspect, one possible implementation method also includes: Display the operation recording control and receive a second confirmation operation input to the operation recording control; The operator is prompted to scan the target area based on the scanning guidance information; Record and store the operator's relevant operating steps; the relevant operating steps include the scanning process of the target area and the operator's operation process on the electronic device.

[0020] In conjunction with the first aspect, in one possible implementation, the method further includes: Display playback options and receive third confirmation from the operator regarding the playback options; The relevant operation steps are replayed based on the third confirmation operation.

[0021] In conjunction with the first aspect, in one possible implementation, acquiring real-time ultrasound images of the target area of ​​the object under test includes: The electronic device displays scan guidance information, which instructs the operator to scan the target area of ​​the object to be tested to obtain real-time ultrasound images. The scan guidance information includes: standard scan operation and standard scan images corresponding to the target area.

[0022] In conjunction with the first aspect, in one possible implementation where there are multiple target locations, the scanning guidance information also includes the scanning order of the multiple target locations.

[0023] Secondly, this application also provides an electronic device, comprising: A transmission component that connects to an ultrasound device so that electronic devices can receive real-time ultrasound images transmitted by the ultrasound device; The display unit includes a display module and a touch module. The display module is used to display real-time ultrasound images, and the touch module is used to receive user input. The processor is electrically connected to the transmission components and the display unit; A memory electrically connected to a processor, the memory storing a computer program that can run on the processor, wherein when the processor executes the program, it implements the method of any one of the first aspects described above.

[0024] Thirdly, this application also provides a computer program product containing instructions that, when executed, perform any of the methods described in the first aspect above. Attached Figure Description

[0025] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic diagram of the electronic device structure provided in the embodiments of this application; Figure 2 This is a schematic flowchart of a method for processing varicose vein data provided in an embodiment of this application; Figure 3 This is a diagram illustrating the scanning guidance information for the user interface. Figure 4 This is a schematic diagram of the capture function controls in the display interface; Figure 5 This is a schematic diagram of the interface for the first prompt information provided in an embodiment of this application; Figure 6 This is a schematic diagram of the interface for the second prompt information provided in an embodiment of this application; Figure 7 A schematic diagram showing the mapping position of a key point of a vein in a real-time optical image; Figure 8 For the operator to drag and drop Figure 7 A schematic diagram after the mapping position has been corrected; Figure 9 A schematic diagram of markers created at multiple key points of the vein; Figure 10 For based on Figure 9 A schematic diagram of the ligation point location output by the marker; Figure 11 This is a schematic diagram of the reflux indicator label provided in the embodiments of this application; Figure 12 A flowchart illustrating the process of determining key vein points and obtaining reference information for key vein points based on real-time ultrasound imaging, as provided in this application embodiment. Figure 13 A flowchart illustrating the backflow indication labeling process provided in this application embodiment; Figure 14 This is a flowchart illustrating the mapping of venous reflux points and venous return points to real-time optical images, provided as an embodiment of this application. Detailed Implementation

[0026] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0027] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The present application will now be described in detail with reference to the accompanying drawings and embodiments. Furthermore, the term "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The terms "first" and "second," etc., in the specification and claims of the embodiments of this application are used to distinguish different objects, not to describe a specific order of objects.

[0028] The following explains the terms used in this application: Escape Point (EP, also known as venous reflux point): refers to the initial anatomical location where blood abnormally shuns from the deep venous system to the superficial venous system during a venous reflux event, i.e., the pressure source of the pathological reflux path.

[0029] Re-entry point (RP, also known as the venous return point): refers to the anatomical location where refluxed blood returns from the superficial venous system to the deep venous system during a venous reflux event.

[0030] Source vein segment: In the venous reflux path, the venous network level where blood flows before a certain venous reflux point / venous return point, which is divided into N1 (deep venous system), N2 (saphenous vein trunk system) or N3 (superficial tributary venous system) according to the hemodynamic stratification system.

[0031] Target venous segment: In the venous reflux path, the venous network level into which blood enters after passing through a certain venous reflux point / venous return point. According to the hemodynamic stratification system, it is divided into N1 (deep venous system), N2 (saphenous vein trunk system) or N3 (superficial tributary venous system).

[0032] Venous shunt: In the pathological state of varicose veins, blood abnormally refluxes from the deep venous system into the superficial venous system through incompetent venous valves, spreads along the superficial venous network, and finally re-enters the deep venous system through perforating veins or functional venous return points, forming a complete hemodynamically abnormal closed loop. A venous shunt event is a core concept describing the venous reflux pathway in the lower extremities. A venous shunt event consists of at least one venous reflux point (reflux initiation point) and at least one venous return point (reflux endpoint), essentially representing a pathological blood flow circulation loop.

[0033] Venous reflux refers to a hemodynamic abnormality in the venous system where blood flows backward in the normal centripetal direction due to venous valve insufficiency or venous wall dilation. In the ultrasound evaluation of varicose veins in the lower extremities, venous reflux is manifested as the reversal of the color of the blood flow patch and the reversal of the blood flow direction on color Doppler images. Reflux lasting longer than a clinically defined threshold (usually 0.5 or 1.0 second) is considered pathological reflux with hemodynamic significance. Venous reflux is a fundamental hemodynamic unit constituting venous shunt events.

[0034] In clinical practice, there are various treatment options for varicose veins, such as vein stripping, laser therapy, radiofrequency ablation, foam sclerotherapy, and targeted ligation. Among these options, targeted ligation is a safe choice that reduces postoperative swelling and lowers the risk of potential complications because it preserves the saphenous vein through targeted ligation rather than completely stripping or closing it.

[0035] Targeted ligation requires surgeons to analyze the hemodynamics and individualized venography of the patient's leg using ultrasound scans. Specifically, surgeons need to create a venous system diagram and identify the locations of reflux points where abnormal blood flow occurs, as well as the return points where the refluxed blood typically flows into the deep veins. To mentally record this crucial information, surgeons often use a waterproof pen to make anatomical markings directly on the patient's leg.

[0036] In recent years, artificial intelligence and image processing technologies have been increasingly applied in the field of medical imaging, aiming to assist clinicians in image analysis, measurement, and improving workflow efficiency. These technologies typically involve classical image processing methods, machine learning algorithms, or deep learning models trained on medical image datasets. Despite these advances, existing technologies often aim to solve isolated technical problems, such as vein segmentation or ultrasound speckle suppression, rather than supporting complex, multi-stage clinical approaches like targeted ligation. Furthermore, these technologies neither integrate domain knowledge for hemodynamic reasoning nor provide a structured representation of functional relationships within the venous system, i.e., vein classification. Therefore, a gap exists between low-level image analysis capabilities and the high-level clinical reasoning required for targeted ligation. Existing software-based systems can only indirectly simplify screening workflows, providing guidance, automation, system storage, or decision support solutions during preoperative screening and diagnosis.

[0037] Among them, existing ultrasound examination software uses artificial intelligence to assist users in performing vascular ultrasound examinations, especially in deep vein thrombosis screening. Its workflow is divided into three steps: the user performs a scan under the guidance of real-time artificial intelligence to determine the position of the ultrasound probe and complete the anatomical structure segmentation, and uploads the segmented key images to the cloud dashboard, where remote radiologists or clinicians review the images and issue reports. The system reduces the professional threshold for ultrasound examinations based on the guidance function and anatomical structure segmentation function.

[0038] Existing enhanced vascular analysis technology imports completed vascular ultrasound videos and automatically analyzes them to generate structured clinical reports. It is particularly suitable for carotid artery screening, and uses deep learning, text recognition and signal processing technologies to interpret single ultrasound images to generate complete ultrasound screening reports.

[0039] Existing report management software only provides document recording and reporting tools for ultrasound data. It receives data collected and measured by ultrasound technicians and organizes it into standardized reports. It does not provide any guidance during the scanning process or provide real-time feedback to surgeons. It is only used as a vascular reporting software platform to collect and record lower extremity venous examination data for longitudinal case review.

[0040] Existing non-invasive, non-thermal varicose vein treatment devices utilize acoustic cavitation to generate high-intensity focused ultrasound. This creates microbubbles within the vein wall, and the rupture of these bubbles mechanically erodes the endothelial cells, ultimately leading to fibrosis and occlusion of the vein. The entire treatment process requires no injections, heating, or surgical incisions. Before and during treatment, the device uses computer vision and artificial intelligence safety features to identify the target vein. It can be applied to the clinical treatment of superficial vein diseases of the lower extremities and achieves automatic vein detection through artificial intelligence analysis of ultrasound video.

[0041] Specifically, the existing technology has the following drawbacks: First, the cognitive load is high and the examination time is long: Ultrasound screening and diagnosis require the simultaneous management of multiple tasks, such as probe positioning, patient positioning, coordination of Vassarva movements, Doppler parameter adjustment, and real-time hemodynamic interpretation. This not only reduces the repeatability of the examination and prolongs the examination time for a single patient, limiting its scalability in different clinical environments, but also requires surgeons to have in-depth knowledge of venous hemodynamics and superb ultrasound skills. For example, they need to independently determine the scanning sequence, select the probe position and direction, switch imaging modes, and judge the timing and location of color Doppler reflux measurement. This further increases the cognitive load, resulting in variability in examination quality, which is more pronounced in less experienced surgeons.

[0042] Second, the system lacks an automatic reflux detection function: the current system requires clinicians to manually interpret color Doppler and pulse wave Doppler signals to determine the presence and duration of venous reflux. This process is subjective and highly dependent on the surgeon's professional skills (such as ultrasound probe positioning), which can easily lead to errors in reflux interpretation. Currently, there is no software that can classify reflux events or assess their hemodynamic significance based on color Doppler images through automatic spatiotemporal analysis.

[0043] Third, it lacks automatic vein classification and tracking functions: It cannot automatically classify vein structures based on the anatomical or functional type of ultrasound images in ultrasound screening, nor can it maintain the spatial continuity of identified vein segments under multiple probe positions and multiple acquisition segments in a single screening. Surgeons need to manually identify veins and integrate them into a coherent venous network in their minds. This process requires professional anatomical knowledge and is prone to misjudgment and spatial correlation errors.

[0044] Fourth, the lack of automatic shunt detection and classification functions makes it impossible to automatically identify venous shunts from ultrasound screening. The identification and classification of shunt types directly determine the ligation strategy. This process requires extensive professional training and relies entirely on manual clinical reasoning, resulting in a long training cycle, inconsistent diagnostic results, and inexperienced surgeons being prone to developing incorrect treatment plans.

[0045] Fifth, the lack of automated patient-specific venous atlas drawing functions and standardized recording and visualization mechanisms: Currently, ultrasound screening relies on the comprehensive analysis of multiple landmarks such as the location of venous reflux points, venous return points, the relationship between superficial and deep veins, and shunt pathways. However, this information is often recorded outside the ultrasound system (e.g., handwritten notes, static charts), disconnected from the original images, easily leading to missing information and making subsequent follow-up review difficult. Furthermore, due to the lack of automated patient-specific venous atlas drawing functions and standardized recording and visualization mechanisms, data from existing reporting platforms requires manual input, cannot be effectively correlated with ultrasound screening results and different examination time periods, and the generated documents are incomplete, non-standardized, and difficult to use for longitudinal retrospective analysis.

[0046] Sixth, the lack of standardized scanning guidelines makes it impossible to provide step-by-step guidance for biplane ultrasound scanning. The screening process relies entirely on the knowledge and experience of surgeons, which may lead to significant differences among operators, incomplete examinations, or even the omission of venous reflux points or venous return points.

[0047] Based on this, embodiments of this application provide a method, electronic device, and product for processing varicose vein data, which can realize a complete closed loop of the targeted ligation preoperative assessment process from ultrasound scanning, key point identification, shunt type determination to ligation point determination, reducing reliance on the surgeon's venous expertise and ultrasound operation skills, reducing examination deviations caused by differences in operator experience, and providing standardized and traceable data support for review and follow-up comparison by mapping key points to real-time optical images of preset sites.

[0048] The method provided in this application embodiment can be applied to, for example... Figure 1 The electronic device 10 shown includes: The transmission component 101 can be connected to an ultrasound device so that the electronic device 10 can receive real-time ultrasound images transmitted by the ultrasound device. The display unit 104 includes a display module 1041 and a touch module 1042. The display module 1041 is used to display real-time ultrasound images, and the touch module 1042 is used to receive user input. The processor 103 is electrically connected to the transmission component 101 and the display unit 104; The memory 102 is electrically connected to the processor 103. The memory 102 stores a computer program that can run on the processor 103. When the processor 103 executes the program, it implements the varicose vein data processing method provided in this application.

[0049] Specifically, the electronic device 10 can be any one of a mobile phone, laptop, desktop computer, or tablet computer; The display unit 104 is used to present real-time ultrasound images and visual information such as vein key point markers, shunt indicators and reflux indicators superimposed on them to the operator. At the same time, it receives the operator's control commands through input devices such as touch screen, mouse or keyboard, including confirmation operations of function controls and drag operations on the display page.

[0050] In one possible implementation, the transmission component 101 is a video capture card, which is connected to the HDMI output port of the ultrasound device to receive real-time ultrasound images output by the ultrasound device; the video capture card can interface with any commercial ultrasound device equipped with an HDMI output port.

[0051] In addition, the processor 103 is also configured to import ultrasound video files in MP4 or DICOM format, thereby enabling offline analysis workflows after acquisition on the same software platform, such as identification of key venous points, determination of venous shunt types, and recommendation of ligation points.

[0052] In another possible implementation, the transmission component 101 also includes an acquisition adapter, one end of which is matched with the DisplayPort, DVI, or composite video output port of the ultrasound device, and the other end is connected to a video acquisition card to adapt to ultrasound devices with different video output interfaces and complete the acquisition of real-time ultrasound images.

[0053] Furthermore, when the ultrasound device supports video streaming via a local wired or wireless network, the transmission component 101 is a wired network interface or a wireless network module, and the display unit 104 directly receives the real-time ultrasound images pushed by the ultrasound device through the wired network interface or wireless network module.

[0054] In another possible implementation, when the electronic device 10 is running the computer program of this application and the imaging software of the ultrasound device at the same time, the transmission component 101 is a screen capture program, which directly captures the display screen in the ultrasound imaging software window as a real-time ultrasound image.

[0055] The computer program runs on any of the following operating systems: Windows, macOS, or Linux. It is designed to assist both professional and non-professional operators in performing standardized biplane ultrasound examinations, covering the entire process from guided scanning to venous shunt classification and ligation point recommendation. Notably, the computer program is an offline program, requiring no network or internet connection. In other words, all data processing and local storage operations of the computer program can be completed on the electronic device 10, without relying on specific hardware. Deployed as a locally hosted web application, the computer program runs entirely on the electronic device 10 without an internet connection and can be accessed through a local browser. Alternatively, the computer program can be deployed on a tablet device equipped with a connected capture card and a webcam, thereby achieving a more portable examination setup without changing the core system architecture. Furthermore, storage and / or video processing can be placed in the cloud, a configuration requiring an internet connection.

[0056] The ultrasound equipment is a portable and desktop dual-function ultrasound imaging system that is widely used in clinical practice for venous examinations and can provide B-mode imaging, color Doppler, and pulse wave Doppler functions.

[0057] Furthermore, the electronic device 10 also includes an optical webcam to enhance the optical tracking capability of the ultrasound probe; the optical webcam can be built into the electronic device 10 or connected externally to the electronic device 10 via USB and connected to the processor 103; when the optical webcam is activated, a continuous optical video stream is captured, and the processor 103 processes the continuous optical video stream to detect the physical position of the ultrasound probe relative to the limb of the object under test within the field of view of the optical webcam.

[0058] like Figure 2As shown, this application also provides a method for processing varicose vein data, the method comprising: S201. Acquire real-time ultrasound images of the target area of ​​the subject to be tested, determine the key vein points in the target area based on the real-time ultrasound images, and obtain reference information of the key vein points; the reference information includes the type of the key vein point, the relevant vein segment, the relevant limb segment, and the location information. The types of key vein points include venous reflux points and venous return points. Taking lower extremity varicose veins as an example, the target site is any anatomical location where varicose veins may occur, which can be at least one of the following: the saphenofemoral junction (SFJ), the great saphenous vein high segment (GSV thigh), the great saphenous vein calf segment (GSVcalf), the saphenopopliteal junction (SPJ), and the small saphenous vein (SSV). Of course, this application can be applied to the identification of ultrasound images of any location, indicating ligation points for any varicose vein site; this application does not limit this.

[0059] Real-time ultrasound images can be at least one of B-mode ultrasound, color Doppler, or pulsed Doppler (PWD).

[0060] Reference information refers to relevant parameters used to assist in hemodynamic analysis. For example, the functional role of a venous key point in the reflux path can be determined based on its type; the venous network hierarchy before and after blood flows through the point can be determined based on the relevant venous segment; the macroscopic anatomical segment of the point in the long axis of the lower limb can be determined based on the relevant limb segment; and the spatial mapping coordinates of the point on the visual reference map of the target site can be determined based on the location information.

[0061] The relevant venous segment refers to the source venous segment and the target venous segment of a certain venous key point. The source venous segment is the venous network level upstream of the venous key point, and the target venous segment is the venous network level downstream of the venous key point. The source venous segment and the target venous segment belong to a certain level of N1, N2 or N3 respectively.

[0062] The relevant limb segment refers to the anatomical area where a venous key point is located. Taking the venous key point of the lower limb as an example, its relevant limb segment can be at least one of the following: hip joint to knee joint, knee joint to ankle joint, or hip joint to ankle joint.

[0063] It should be noted that the hemodynamic abnormalities causing varicose veins can be termed venous shunt events. A venous shunt event comprises a complete pathological reflux closure loop, which is formed by at least one venous reflux point and at least one venous return point. However, there is not a one-to-one correspondence between venous reflux points and venous return points. For example, a venous reflux point may correspond to multiple venous return points due to branches of its downstream superficial venous network, or multiple independent venous reflux points may each correspond to their respective venous return points.

[0064] Location information refers to the spatial mapping position of key vein points in a real-time optical image generated based on the target site. In one possible implementation, the location information is obtained by an inertial measurement unit (IMU) sensor mounted on the electronic device 10. The real-time optical image is displayed on the user interface of the electronic device 10 and is obtained using an optical webcam mounted on the electronic device 10. For example, the real-time optical image can be an anatomical diagram of the target site, a preset photograph of a standard human lower limb template, a real photograph of the target site of the subject under test, or a three-dimensional surface mesh model of the target site. The three-dimensional surface mesh model contains the three-dimensional spatial location information of each point on the surface of the target site and can be reconstructed by scanning with a depth camera or structured light sensor mounted on the electronic device 10.

[0065] One possible implementation involves acquiring real-time ultrasound images of the target area of ​​the object under test, including: The electronic device 10 displays scanning guidance information, which instructs the operator to scan the target area of ​​the object to be tested to obtain real-time ultrasound images. The scanning guidance information includes: standard scanning operation and standard scan images corresponding to the target area.

[0066] In one possible implementation, the standard operation audio-visual information is at least one of the following: a pre-recorded standard operation video, multiple preset standard operation images, or pre-recorded standard operation audio. The standard operation images are annotated anatomical diagrams with text descriptions, used to visually demonstrate to the operator the position of the ultrasound probe and the expected anatomical structures for the current scanning step. The standard operation audio can be provided synchronously with the standard operation images; or it can be configured independently of the standard operation images and played synchronously with the real-time ultrasound images, providing voice guidance prompts.

[0067] The following example illustrates the standard operation video for a specific target area: Two pre-recorded video clips stored locally on electronic device 10 are played: a standard scanning operation and a standard scan image. For example... Figure 3In the displayed interface, the video clip at the top is a standard scanning operation, demonstrating the correct placement and orientation of the ultrasound probe at the target site. The video clip below is a standard scan image, which refers to the expected standard ultrasound image, corresponding to the anatomical location in the video clip above.

[0068] These two pre-recorded video clips serve as visual guides to help operators correctly position the ultrasound probe and identify the expected ultrasound presentation. Notably, this scan guidance information is displayed synchronously with the real-time ultrasound images transmitted from the video capture card, allowing operators to compare the expected ultrasound images with the real-time ultrasound images on the same interface.

[0069] It is worth mentioning that if there are multiple target areas, the scanning guidance information will also include the scanning order of the multiple target areas.

[0070] For example, such as Figure 3 As shown in the left column, the scanning sequence is presented in the form of an operation checklist, which includes five steps, each of which is a standard anatomical region to be scanned and screened: 1. Saphenofemoral junction (SFJ), 2. Great saphenous vein thigh segment (GSV thigh), 3. Great saphenous vein calf segment (GSV calf), 4. Saphenofemoral junction (SPJ), and 5. Small saphenous vein (SSV).

[0071] Furthermore, the scan guidance information also includes extension files, which allow operators to customize or extend the preset scan steps according to the specific specifications of their institution to accommodate additional anatomical areas or personalized scan protocols.

[0072] This application provides standardized ultrasound screening procedures through scan guidance information, accompanied by standard ultrasound images and probe positioning demonstration videos. This improves the completeness, accuracy, and consistency of ultrasound scans performed by different operators, while reducing reliance on operator expertise, shortening training time for novice operators, and minimizing the possibility of missed anatomical areas during ultrasound acquisition.

[0073] S202. Identify venous shunts at the target site based on reference information of key venous points, and determine the shunt type of the venous shunt, which is the abnormal blood flow that causes varicose veins. In one possible implementation, the shunt type is the shunt type under the Conservative Hemodynamic Cure of Venous Insufficiency (CHIVA) technique.

[0074] One possible implementation involves determining the shunt type of venous shunt based on reference information from key venous points, specifically as follows: Determine the blood flow path of venous shunts based on reference information from key venous points; The classification rules are obtained, and the shunt type of venous shunt is determined based on the classification rules. The classification rules include multiple shunt types and the blood flow path corresponding to each shunt type.

[0075] One possible implementation method is to use a decision tree to identify the traffic splitting event and the traffic splitting type. The specific judgment logic is as follows: For Type I shunt (Shunt I) First, check if there is an EP in the saphenous-femoral junction (SFJ) region, with the blood flow path being N1→N2; if it is present, continue to check if there is a RP in the main saphenous vein (N2) region, with the blood flow path being N2→N1; if no EP or RP related to the tributary vein (N3) is detected throughout the entire process, and both core conditions are met, it is determined to be a type I shunt.

[0076] For Type II shunt (Shunt II) First, check whether there is an EP in the area of ​​the main saphenous vein (N2) or its tributary vein (N3), with the blood flow path being N2→N3; if it exists, and no RP that can return to the deep vein (N1) is detected, and N1 closed loop return cannot be formed, then it is determined to be a type II shunt.

[0077] For Type III shunt (Shunt III) First, EP is detected to correspond to the blood flow paths N1→N2 and N2→N3 in sequence; if RP is detected in the tributary vein (N3) area at the same time, the blood flow path is N3→N1, forming a complete blood flow closed loop, then it is determined to be a type III shunt.

[0078] For Type IV shunt (Shunt IV) First, check if there is an EP in the connection area between the pelvic venous plexus and the main saphenous vein (N2), with the blood flow path being pelvic venous plexus → N2; if it exists, continue to check if there is an RP in the area of ​​the main saphenous vein (N2), with the blood flow path being N2 → N1, forming a closed-loop blood flow, which is then determined to be a type IV shunt.

[0079] For Type V shunt (Shunt V) First, the blood flow path corresponding to EP is detected as N1→N2→N3, and the reflux is traced back to the pelvic venous plexus. If RP is detected in the area of ​​the tributary vein (N3) at the same time, the blood flow path is N3→N1, forming a closed blood flow loop, then it is determined to be a type V shunt.

[0080] For Type VI shunt (Shunt VI) First, for EP, which corresponds to a continuous blood flow path from the pelvic venous plexus or perforating veins directly to the collateral vein (N3), the reflux does not involve the saphenous vein (N2); when the reflux point RP can return blood flow to the deep vein (N1) through another perforating vein or saphenous vein (N2), it is determined to be a type VI shunt.

[0081] All other decisions related to shunt are performed in accordance with the decision tree rules mentioned above, that is, to determine whether there are entry points (EPs) and reflux points (RPs) that conform to the blood flow direction in the specified anatomical location and corresponding venous segment.

[0082] The classification rules can be based on general standard guidelines for the treatment of varicose veins. For example, the classification rules are shown in Table 1: Table 1. Classification Rules and Recommended Ligation Points

[0083] S203. Determine and indicate the ligation point of the target site based on the shunt type.

[0084] One possible implementation is that the built-in judgment logic in the rule engine also includes the ligation point corresponding to each diversion type. The correspondence between diversion types and ligation points is shown in Table 1.

[0085] After the triage type is determined, the rule engine outputs the ligation point based on the correspondence between the type and the ligation point.

[0086] This application employs a decision support algorithm to first classify detected venous shunts into clinically significant shunt types based on the location and connectivity of key venous points. Subsequently, ligation points are determined according to the classified shunt structures, supporting targeted ligation while maintaining normal venous return pathways. This improves interpretability, ensures consistency in clinical decision-making, and helps enhance the efficiency of operator training.

[0087] One possible implementation is that when the classification rules built into the rule engine of this application differ from the internal treatment guidelines of a specific hospital or medical institution, a locally installed application programming interface can be called to connect to the institution's own clinical decision support system. The latter can then perform a secondary analysis of the venous key points and their reference information in the system according to its specific internal rules, update them to obtain the triage type and ligation point that conforms to the institution's personalized guidelines, and return the updated triage type and ligation point to the system of this application. This ensures that the final treatment plan strictly follows the institution's own treatment guidelines, rather than mechanically applying general standards.

[0088] This application enables the identification of venous shunt events and determination of shunt type based on real-time ultrasound imaging, thereby reducing reliance on the operator's expertise in venous hemodynamics and improving the consistency of diagnostic results among different operators and in different clinical settings. Furthermore, this application also determines and outputs the corresponding ligation point based on the shunt type, thus providing a traceable and repeatable treatment plan.

[0089] One possible implementation is, such as Figure 12 As shown, key vein points in the target area are determined based on real-time ultrasound images, and reference information for these key vein points is obtained, including: S1201. In real-time ultrasound images, shunt indicators and reflux indicators are marked; shunt indicators are used to characterize abnormal venous proximity, and reflux indicators are used to characterize venous reflux; One possible implementation involves annotating shunt indications on real-time ultrasound images, specifically including: The distribution of veins in real-time ultrasound images is identified. When the spatial distance between any two veins is less than a preset threshold, a shunt indicator is marked at a preset position in the real-time ultrasound image.

[0090] Specifically, identifying the distribution of veins in real-time ultrasound images involves: The distribution information of each vein in the real-time ultrasound image is obtained. The distribution information of each vein is obtained by the operator through annotation or determined based on a pre-trained first model. The input of the first model is the real-time ultrasound image, and the output is the distribution information of each vein in the real-time ultrasound image.

[0091] The preset location can be an abnormally close venous intersection.

[0092] One possible implementation is that the first pre-trained model can be a medical image segmentation and classification algorithm to identify venous structures in ultrasound images. Then, a heuristic algorithm is used to classify each detected vein into the correct anatomical category, such as the great saphenous vein (GSV), small saphenous vein (SSV), branch veins, perforating veins, or deep veins. When the minimum spatial distance between the lumens of any two veins of different categories is less than a preset threshold, a shunt indication is marked at the abnormally close intersection of the two veins in the real-time ultrasound image.

[0093] For example, the Segment Anything Model 2 (SM2) image and video segmentation model can be used to segment the vein lumen in real-time ultrasound images and extract the lumen contour and spatial coordinates of each vein. Then, a heuristic algorithm is used to classify the segmented veins into anatomical categories.

[0094] The heuristic algorithm, based on clinical knowledge of lower extremity venous anatomy, pre-sets the following classification rules: First, based on the anatomical location and accompanying relationship of veins, a core classification is made, and veins located deep in the muscle fascia and accompanied by the same-named artery are classified as deep veins; Secondly, veins located in the double fascial space between the saphenous vein fascia and the muscle fascia, exhibiting a typical "eye sign" in ultrasound cross-section, and whose course conforms to the fixed anatomical area of ​​the saphenous vein trunk, are classified as the great saphenous vein or the small saphenous vein trunk. Furthermore, venous segments that penetrate the deep fascia in spatial location and directly connect the identified superficial and deep venous systems are classified as perforating veins. Finally, veins with a lumen diameter smaller than the corresponding saphenous vein trunk, which are only surrounded by skin and superficial fascia and do not have the characteristics of double fascial wrapping, and which flow into the corresponding saphenous vein trunk, are classified as tributary veins or branch veins.

[0095] Furthermore, after completing vein segmentation and classification, the minimum lumen spacing of different types of veins is calculated based on spatial coordinates. When the spacing is less than a preset threshold, a diversion indication is marked.

[0096] This application automatically classifies venous structures and detects the spatial proximity between two classified venous lumens in real time in ultrasound images. This establishes a tracking relationship for the same venous segments under different probe physical positions, thereby stitching together discrete local sections to restore a complete venous network topology and ensuring the spatial continuity of identified venous segments during the detection process.

[0097] In other possible implementations, the operator can be prompted to mark each vein on the screen, and the system analyzes the marked area according to a classification algorithm to classify each vein. Specifically, the classification algorithm can be a traditional image processing technique (such as a tubular structure detection algorithm or an intensity-based segmentation algorithm). When the classification confidence based on real-time ultrasound images is low, classification can be performed based solely on the anatomical location image of the target site and the ultrasound probe position, which are obtained from an optical webcam.

[0098] This application automatically identifies potential venous shunts by analyzing the spatiotemporal patterns of ultrasound and the anatomical relationships within the superficial and deep venous systems. This helps reduce the omission of inconspicuous shunts during the detection process, thereby improving diagnostic sensitivity while reducing the cognitive load on the operator.

[0099] The following explains the labeling process for reflux indicators: One possible implementation is that the reflux indication includes multiple reflux indications with different warning levels, such as... Figure 13 As shown, the regurgitation indication is marked on the real-time ultrasound image, specifically as follows: S1301. Identify multiple venous refluxes in real-time ultrasound images and obtain the reflux time and reflux area of ​​each venous reflux. When the real-time ultrasound image is a color Doppler image, the reflux area refers to the venous lumen area in the color Doppler image that is confirmed to have pathological reflux after pixel-level blood flow direction analysis.

[0100] One possible implementation, when the real-time ultrasound image is a color Doppler image, is as follows: S1301 can be: identifying venous reflux using traditional computer vision methods, specifically, detecting based on pixel color groups in the color Doppler image, where blue gradients correspond to negative blood flow and yellow-red gradients correspond to positive blood flow, thereby locating the reflux region; to avoid noise interference, using region size as a filtering condition, ignoring tiny regions with an area smaller than a preset threshold; for larger reflux regions, tracking the same reflux region in consecutive frames using a region overlap tracking method, thereby measuring the duration of reflux.

[0101] The region overlap tracking method is as follows: In consecutive color Doppler image frames, calculate the overlap area ratio between each reflux region in the current frame and all reflux regions in the previous frame; when the overlap area ratio exceeds a preset threshold, it is determined that the same reflux region is a continuation in different frames; continuously track the reflux region until it disappears in the image, record the total number of frames from the first appearance of the region to its final disappearance, and calculate the duration of the venous reflux by combining the time interval of a single frame image.

[0102] This application improves examination efficiency by directly detecting venous reflux time and marking reflux indicators from real-time ultrasound images, thus eliminating the need for additional manual interpretation of pulse wave Doppler data.

[0103] In another possible implementation, S1301 is specifically as follows: Identify the reflux region of each venous reflux in real-time ultrasound images; In each reflux region of venous reflux, a sampling point is selected to be analyzed. Specifically, the sampling point refers to the sampling location selected in the center of the reflux region, which is used to quantify the reflux time. For example, the sampling point can be the center coordinate of the lumen where the reflux signal is most stable and the color block boundary is clearest.

[0104] Pulse Doppler waveform analysis was performed on each point to be analyzed to obtain the reflux time of each venous reflux.

[0105] In one possible implementation, after a potential reflux event is automatically detected, the suspected reflux area is first marked in the real-time ultrasound image, and then the operator manually confirms or excludes the mark to decide whether to record it as a valid discovery.

[0106] S1302. For each venous reflux, based on the reflux time, determine the warning level of venous reflux, and mark the reflux indication corresponding to the warning level in the reflux area of ​​venous reflux.

[0107] One possible implementation is to use 0.5 seconds as the threshold for judging the first level of warning; and 0.5 seconds as the threshold for judging the second level of warning.

[0108] It is worth noting that the judgment thresholds are based on clinically established criteria for significant hemodynamic reflux. This application allows operators to configure different values ​​for the judgment thresholds at each level to adapt to different clinical standards or institutional protocols.

[0109] One possible implementation is, such as Figure 11 As shown, the backflow indication is represented by different visual warning signs, superimposed on the corresponding backflow area.

[0110] In other embodiments, reflux indicators may also be color-coded visual warnings, for example, a yellow warning may be superimposed when the reflux time exceeds 0.5 seconds; a red warning may be superimposed when the reflux time exceeds 1 second in the positive blood flow direction; and a dark blue indicator may be superimposed when the reflux time exceeds 1 second in the negative blood flow direction.

[0111] It is worth mentioning that, in the example corresponding to the color-coded visual warning, under the same reflux time, this application also sets different colors of reflux indicators based on the blood flow direction, so that under the same reflux severity, the operator can quickly distinguish the anatomical source of reflux and the direction of blood flow, thereby improving scanning efficiency and judgment accuracy.

[0112] One possible implementation includes, before identifying the distribution of veins in real-time ultrasound images: For real-time ultrasound images, the system displays classification options for imaging modes and identifies the imaging mode based on the operator's selection of classification options; or it identifies the imaging mode of the real-time ultrasound image based on a pre-trained classification model.

[0113] Specifically, the pre-trained classification model identifies the brand of ultrasound equipment through video frames of real-time ultrasound images and determines the imaging mode based on the brand.

[0114] In one possible implementation, the classification model is an ultrasound device and imaging pattern classifier (such as any one of the MobileNetV3 lightweight CNN classifier, ResNet-18 residual network classifier, or SVM support vector machine classifier). The imaging pattern of real-time ultrasound images is then identified based on the pre-trained classification model, specifically as follows: The classifier predicts the model of the ultrasound equipment and the current imaging mode based on the input real-time ultrasound image frames. The imaging mode identification result can be any one of B-mode ultrasound, color Doppler, or pulsed Doppler (PWD).

[0115] The imaging mode is identified based on the operator's selection of classification options. Specifically, the operator manually selects the ultrasound equipment brand and imaging mode through the user interface; detection is performed based on frame color histogram analysis rather than a trained machine learning classifier, utilizing specific features of different imaging modes; and integration with the ultrasound equipment's metadata (if such metadata is contained in the video signal).

[0116] Cropping of real-time ultrasound images based on imaging modes.

[0117] This application identifies imaging patterns to crop regions of interest from image frames and selectively routes the cropped frames to corresponding subsequent processing algorithms, thereby avoiding unnecessary computation and improving overall efficiency.

[0118] It is worth mentioning that this application processes real-time ultrasound images through parallel-running display and processing threads; Specifically, the display thread continuously captures raw video frames of real-time ultrasound images and renders them to the user interface in real time; The processing thread synchronously captures raw video frames and passes them to the image analysis algorithm for analysis and processing; the output of the image analysis algorithm is asynchronously returned to the user interface for overlay visualization on the real-time display screen; the analysis and processing includes annotation of diversion indicators and reflow indicators.

[0119] The display thread of this application is independent of image analysis and processing operations, which can ensure that real-time ultrasound images are displayed to surgeons with low latency.

[0120] This application automates tasks such as vein identification, reflux time measurement, and result recording during the screening process, reducing the cognitive burden on operators who need to simultaneously manage probe positioning, patient positioning, and Doppler parameter adjustments. This shortens examination time and improves scalability in different clinical environments.

[0121] S1202, Receive the first confirmation operation input by the operator based on the shunt indication and reflux indication; the first confirmation operation is used to mark the venous reflux point and venous return point on the real-time ultrasound image; In one possible implementation, receiving the first confirmation operation specifically involves: displaying the capture function control and receiving the user's first confirmation operation on the capture function control.

[0122] like Figure 4In the illustrated embodiment, during the screening process targeting a specific area, when the operator identifies a clinically significant venous reflux point or venous return point on a real-time ultrasound image, the operator needs to perform a first confirmation operation on the capture function control – “Fast Capture” – to activate the “Fast Capture” function in order to record the identified key venous point.

[0123] S1203. Obtain the type of vein key points, related vein segments, related limb segments, and location information.

[0124] One possible implementation involves obtaining the location information of key vein points, including: The venous reflux points and venous return points marked on real-time ultrasound images are mapped to real-time optical images of the target site to obtain the location information of the venous reflux points and venous return points.

[0125] One possible implementation is, such as Figure 14 As shown, the real-time optical image mapping of venous reflux points and venous return points marked on real-time ultrasound images to the target site is as follows: S1401. Upon receiving the first confirmation operation, acquire the current position information of the ultrasonic probe at the target location; The current position information of the ultrasonic probe is obtained based on any one of the following methods: optical tracking, electromagnetic tracking, inertial measurement, depth sensing measurement, or operator determination.

[0126] Specifically, optical tracking involves configuring an optical webcam on the electronic device 10 and analyzing the real-time images captured by the optical webcam using computer vision technology to calculate the positioning information of the ultrasonic probe at the target location. Electromagnetic tracking involves installing an electromagnetic tracking system on the ultrasonic probe to directly read the positioning data of the ultrasonic probe at the target location.

[0127] Inertial measurement is used to acquire the attitude and relative position data of an ultrasonic probe by installing an inertial measurement unit on the probe.

[0128] Depth sensing measurement uses a depth camera or structured light sensor to acquire a three-dimensional spatial reference model of the target area, and determines the position of the ultrasonic probe based on this three-dimensional spatial reference model.

[0129] If the operator determines the location, the operator will manually specify the ultrasound probe location by selecting the corresponding area of ​​the ultrasound probe from the predefined anatomical region on the schematic diagram of the target area, thereby replacing the ultrasound probe location method based on continuous spatial tracking listed above.

[0130] Furthermore, optical tracking and inertial measurement data can be fused to improve the positioning accuracy of the ultrasonic probe under varying lighting conditions.

[0131] S1402. Based on the current location information, map the venous reflux point and venous return point to the real-time optical image.

[0132] It is worth mentioning that the venous reflux point and venous return point are detected from real-time ultrasound images at the current position of the ultrasound probe.

[0133] For example, when the target part is the leg and the current position information is determined by optical tracking, and the real-time optical image is a standard leg diagram, steps S1401~S1402 are as follows: An optical network camera captures real-time operational images of the patient's entire leg and the ultrasound probe. First, a deep learning object detection model identifies the patient's leg region and the ultrasound probe region from the real-time operational images, outputting the bounding box coordinates of the leg region and the center coordinates of the ultrasound probe. Then, pose estimation is performed on the detected leg region, extracting key points of the leg bones and constructing a leg skeleton. This leg skeleton is used to standardize and parameterize the leg region, mapping legs of different lengths to a coordinate range of 0-1, where 0 corresponds to the leg root (groin) and 1 corresponds to the leg tip (toe). The position of the ultrasound probe on the leg region is then converted into coordinate values ​​within the 0-1 range. Simultaneously, a standard leg diagram is pre-parameterized according to the same 0-1 coordinate range, establishing a mapping relationship between the real-time operational images and the standard leg diagram. For example, after deep learning processing, the probe position coordinates are obtained as 0.5, which corresponds to the position of the midpoint of the leg on the standard leg diagram. Finally, based on the corresponding coordinates of the ultrasound probe in the standard leg diagram, and combined with the relative offsets of the venous reflux point and venous return point in the real-time ultrasound image relative to the center of the ultrasound probe, the final absolute coordinates of the two types of key points in the standard leg diagram are calculated. Based on these final absolute coordinates, the venous reflux point and venous return point in the real-time ultrasound image can be superimposed and displayed at the corresponding positions on the standard leg diagram.

[0134] One possible implementation, after mapping the venous reflux point and venous return point to real-time optical images based on the current location information, also includes: The system receives dragging operations from the operator on the currently marked venous reflux point or venous return point on the real-time optical image. The dragging operation is used to mark the venous reflux point or venous return point at the position corrected by the operator.

[0135] like Figure 7 The illustrated embodiment maps a key vein point in a real-time optical image based on the current location information. In this embodiment, when the operator finds the mapped position inaccurate, the mapped key vein point can be dragged, such as... Figure 8As shown, this is the corrected position of the vein key point after the drag operation.

[0136] Another possible approach is to map the venous reflux points and venous return points marked on the real-time ultrasound images to the real-time optical images of the target site, specifically as follows: A second prompt message is displayed to guide the operator to drag and map the venous reflux points and venous return points marked on the real-time ultrasound image to the real-time optical image of the target site.

[0137] like Figure 6 In the illustrated embodiment, a second prompt message is displayed on the user interface of the electronic device 10 to remind the user to perform a drag operation.

[0138] Specifically, after receiving the second prompt, the operator determines the location of the key vein points in the real-time optical image by clicking and dragging, and maps the key vein points to a completely accurate location in a manual positioning manner.

[0139] One possible implementation involves obtaining the type of the vein key point, the relevant vein segment, the relevant limb segment, and the location information, specifically: Display the first prompt message, such as Figure 5 As shown, the first prompt information is used to prompt the operator to mark the relevant venous segments, relevant limb segments, and location information of the venous reflux point in the real-time optical image, as well as to mark the relevant venous segments, relevant limb segments, and location information of the venous return point in the real-time optical image.

[0140] One possible approach is to map the lower limbs into three regions (groin or hip-knee joint, knee-ankle joint, femur-knee joint) instead of a continuous spatial diagram when optical webcams are not available and precise positioning is not required. The operator can then simply record which region the key vein points are located in.

[0141] One possible implementation method also includes: In response to the first confirmation operation, real-time ultrasound images and the operator's real-time scanning operation at the target site are recorded; Specifically, such as Figure 4 As shown in the right sidebar, after activating the "Quick Capture" function control, a five-second real-time ultrasound image segment is recorded from the current time and saved on the electronic device 10 so that it can be played back when the operator clicks a marker. Figure 4 In the embodiment shown, the real-time optical image is a schematic diagram of the legs, and Figure 4 It also includes an "overlay" function, which is used to overlay and display vein segmentation results on ultrasound images in real time.

[0142] In one possible implementation, in addition to the "Quick Capture" function control, this application also includes a "Capture Segment" function control. Both function controls will capture a 5-second video segment. The difference is that Quick Capture does not require the user to input attribute information, such as vein segments.

[0143] It is worth mentioning that the recording duration is set by the operator before or during the scanning process; One possible implementation is that when the system detects a potential venous reflux point or reentry point, the system can automatically trigger the "rapid capture" function after operator confirmation; One possible implementation is to record only a single representative static ultrasound image when video storage is limited. One possible implementation is a voice-activated "quick capture" function as an alternative to the capture function controls, thereby enabling hands-free operation during the scanning process.

[0144] Create markers for reference information, recorded real-time ultrasound images, and real-time scanning operations in real-time optical imaging.

[0145] Specifically, based on the location information of key vein points in real-time optical imaging, markers are created for the recorded real-time ultrasound image segments and real-time scanning operations, so that the markers are displayed at the locations corresponding to the key vein points. After the markers are created, the operator can click on the marker on the real-time optical imaging to replay a five-second recording segment of the real-time ultrasound image.

[0146] For example, such as Figure 9 As shown in the figure, "0" marks a venous reflux point; "1" marks a venous return point.

[0147] One possible implementation method includes activating the "assessment" function by receiving confirmation from the operator regarding the assessment options after the operator completes the screening process for the target site and creates relevant markers. Based on all markers on the real-time optical image, reference information for each key venous point is obtained, including its type, relevant venous segment (N1-N3), relevant limb segment, and location information in the real-time optical image. Based on the reference information, the system analyzes the data to determine the shunt type, demonstrates the reasoning behind the classification, and proposes suggested intraoperative ligation points (or multiple ligation points). The assessment results and suggestions are displayed in the user interface and saved.

[0148] like Figure 10 As shown, Figure 9 The illustrated embodiment outputs a schematic diagram of the ligation point after activating the "Evaluation" function. Figure 10 The point indicated by the middle arrow is the ligation point.

[0149] against Figure 9 In the illustrated embodiment, this application also displays the following evaluation results and recommended solutions on the user interface: Evaluation result: Type I shunt was detected.

[0150] Reasoning basis: There is a venous reflux point from N1 to N2 at the junction of the saphenofemoral vein, but no venous reflux point from N2 to N3; there is a venous return point along N2, but no venous return point along N3.

[0151] Recommended ligation procedure (treatment plan): Ligate the junction of the saphenofemoral vein.

[0152] This application visualizes the specific venous network of the subject through real-time optical imaging and records real-time ultrasound images and ultrasound probe scan segments at important anatomical landmarks (such as venous return points and venous reflux points). This eliminates the need for operators to memorize the specific venous diagrams of the subject at different stages of the examination, supports efficient re-examination and longitudinal follow-up, and avoids the problems of traditional handwritten notes or static charts being disconnected from the original image data, information loss, and difficulty in retrospective comparison during follow-up.

[0153] One possible implementation method also includes: Display the operation recording control and receive a second confirmation operation input to the operation recording control; The operator is prompted to scan the target area based on the scanning guidance information; Record and store the operator's relevant operating steps; the relevant operating steps include the scanning process of the target area and the operator's operation process on the electronic device 10.

[0154] The relevant operation steps refer to all the operator's actions during the ultrasound screening and diagnosis process, specifically including: the operator's scanning process at the target site, the real-time ultrasound image observed by the operator on the screen of the electronic device 10, the five-second ultrasound video clip corresponding to the operator activating the capture function to record the venous reflux point or venous return point and the structured annotation information filled in by the operator, the interactive action of the operator manually dragging and adjusting the position of the marker on the leg diagram, the operator's click operation to confirm or exclude potential reflux events automatically marked by the system, and the interface output of the shunt type determination result and ligation suggestion generated by the system after the operator activates the assessment function.

[0155] When the recording function is activated based on the second confirmation operation, a dedicated background recording thread continuously writes the video frames received during the operator's scanning of the target area to the local storage of the electronic device 10. When the electronic device 10 is equipped with an optical webcam, the background recording thread simultaneously and independently writes the ultrasound video stream output by the ultrasound device and the video stream from the optical webcam. When no webcam is connected, the background recording thread only writes the ultrasound video stream. The background recording thread runs independently of the display thread and processing thread mentioned above, thereby ensuring that the recording activity does not interfere with real-time display or real-time analysis and processing.

[0156] Optionally, the recorded operation steps can be encrypted and exported to the offline storage system of the local hospital network; and interface with the locally installed electronic medical record system based on a standard local interface; and for research or auditing purposes, an anonymous version of the session data can be exported in a structured format.

[0157] To reduce storage requirements, one possible implementation method is to record and store the first process of the operator scanning the target area to determine the ligation point information, specifically as follows: The first process is recorded based on a preset recording interval, resulting in multiple target frames of the first process, which are then stored; or, The first process is recorded, and the recording results are stored on an external storage device or cloud storage; or, The first process is recorded, and the recording results are compressed and stored using configurable quality and format compression video encoding to achieve a balance between storage consumption and playback fidelity.

[0158] One possible implementation, the method further includes: Display playback options and receive third confirmation from the operator regarding the playback options; The relevant operation steps are replayed based on the third confirmation operation.

[0159] This application allows operators to access and review complete ultrasound screening records and results. For example, operators can replay the entire recorded video, view individually captured venous key point segments, review annotated real-time optical images, and consult shunt types and ligation recommendations, thereby enabling traceable review of the entire screening process, facilitating intraoperative decision verification and postoperative teaching analysis.

[0160] This invention achieves a closed-loop process for the entire preoperative assessment of targeted ligation, integrating standard scan guidance, real-time ultrasound image analysis, venous key point capture, automatic determination of shunt type, and ligation strategy output into a single system. This avoids information transmission gaps and low process efficiency caused by the separation of screening, recording, analysis, and decision-making processes.

[0161] It should be noted that although the operations of the method of the present invention are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all of the operations shown must be performed to achieve the desired result. On the contrary, the steps depicted in the flowchart may be performed in a different order. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0162] On the other hand, embodiments of this application provide a computer program product including instructions that, when executed, cause the method described in embodiments of this application to be performed. For example, it can execute... Figure 2 The steps of the method shown are as follows.

[0163] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0164] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A method for processing varicose vein data, characterized in that, Applied to electronic devices, the method includes: Real-time ultrasound images of the target area of ​​the subject are acquired, and key venous points in the target area are determined based on the real-time ultrasound images. Reference information of the key venous points is obtained. The reference information includes the type of the key venous point, the relevant venous segment, the relevant limb segment, and the location information. The type of the key venous point includes venous reflux point and venous return point. Based on the reference information of the key venous points, the venous shunt at the target site is identified, and the shunt type is determined, wherein the venous shunt is an abnormal blood flow that causes varicose veins; Based on the shunt type, the ligation point of the target site is determined and indicated.

2. The method according to claim 1, characterized in that, Based on the real-time ultrasound images, key venous points in the target area are determined, and reference information of the key venous points is obtained, including: The real-time ultrasound images are labeled with shunt and reflux indicators; the shunt indicator is used to characterize abnormal venous proximity, and the reflux indicator is used to characterize venous reflux. The system receives a first confirmation operation input from the operator based on the shunt indication and the reflux indication; the first confirmation operation is used to mark the venous reflux point and the venous return point on the real-time ultrasound image. Obtain the type, related vein segment, related limb segment, and location information of the key vein points.

3. The method according to claim 2, characterized in that, Obtaining the location information of the key vein points includes: The venous reflux points and venous return points marked on the real-time ultrasound images are mapped to the real-time optical images of the target site to obtain the location information of the venous reflux points and venous return points.

4. The method according to claim 2, characterized in that, The real-time ultrasound image annotation for shunt indication specifically includes: The distribution of each vein in the real-time ultrasound image is identified. When the spatial distance between any two veins is less than a preset threshold, the shunt indicator is marked at a preset position in the real-time ultrasound image.

5. The method according to claim 4, characterized in that, Identifying the distribution of veins in the real-time ultrasound image, specifically: The distribution information of each vein in the real-time ultrasound image is obtained, wherein the distribution information of each vein is obtained by the operator through annotation, or determined based on a pre-trained first model; the input of the first model is the real-time ultrasound image, and the output is the distribution information of each vein in the real-time ultrasound image.

6. The method according to claim 2, characterized in that, The reflux indication includes multiple reflux indications with different warning levels. The reflux indication is then annotated on the real-time ultrasound image, specifically as follows: Multiple venous refluxes are identified in the real-time ultrasound images, and the reflux time and reflux area of ​​each venous reflux are obtained; For each venous reflux, based on the reflux time, the warning level of the venous reflux is determined, and a reflux indicator corresponding to the warning level is marked in the reflux area of ​​the venous reflux.

7. The method according to claim 6, characterized in that, Multiple venous refluxes were identified in the real-time ultrasound images, specifically: Identify the reflux region of each venous reflux in the real-time ultrasound images; Select the points to be analyzed in each reflux region of the venous reflux. Pulse Doppler waveform analysis was performed on each of the points to be analyzed to obtain the reflux time of each venous reflux.

8. The method according to claim 2, characterized in that, The type, related vein segment, related limb segment, and location information of the key vein points are obtained, specifically as follows: The first prompt message is displayed, which prompts the operator to mark the relevant venous segment, relevant limb segment and location information of the venous reflux point in the real-time optical image, and to mark the relevant venous segment, relevant limb segment and location information of the venous return point in the real-time optical image.

9. The method according to claim 3, characterized in that, Mapping the venous reflux points and venous return points marked on the real-time ultrasound images to the real-time optical images of the target site, specifically: Upon receiving the first confirmation operation, the current position information of the ultrasonic probe at the target location is obtained; Based on the current location information, the venous reflux point and the venous return point are mapped to the real-time optical image.

10. The method according to claim 9, characterized in that: After mapping the venous reflux point and the venous return point to the real-time optical image based on the current location information, the method further includes: The operator receives a drag operation from the operator on the real-time optical image to mark the currently marked venous reflux point or venous return point at the position corrected by the operator.

11. The method according to claim 3, characterized in that, Mapping the venous reflux points and venous return points marked on the real-time ultrasound image to the real-time optical image of the target site, specifically: A second prompt message is displayed to instruct the operator to drag and map the venous reflux points and venous return points marked on the real-time ultrasound image to the real-time optical image of the target site.

12. The method according to claim 2, characterized in that, The method further includes: In response to the first confirmation operation, the real-time ultrasound image and the operator's real-time scanning operation at the target site are recorded; Markers are created in the real-time optical image for the reference information, the recorded real-time ultrasound image, and the real-time scanning operation.

13. The method according to claim 1, characterized in that, The shunt type of venous shunt is determined based on the reference information of the aforementioned key venous points, specifically as follows: The blood flow path of the venous shunt is determined based on the reference information of the key points of the vein; Obtain classification rules and determine the shunt type of the venous shunt based on the classification rules; The classification rules include multiple diversion types and the blood flow path corresponding to each diversion type.

14. The method according to any one of claims 1-13, characterized in that, The method further includes: Display the operation recording control and receive a second confirmation operation input to the operation recording control; The operator is prompted to scan the target area based on the scanning guidance information; The operator's relevant operation steps are recorded and stored; the relevant operation steps include the process of scanning the target area and the operator's operation process on the electronic device.

15. The method according to claim 14, characterized in that, The method further includes: Display playback options and receive a third confirmation from the operator regarding the playback options; The relevant operation steps are replayed based on the third confirmation operation.

16. The method according to any one of claims 1 to 13, characterized in that, Acquire real-time ultrasound images of the target area of ​​the object under test, including: The electronic device displays scanning guidance information, which instructs the operator to scan the target area of ​​the object to be tested to obtain the real-time ultrasound image; the scanning guidance information includes: standard scanning operation and standard scan image corresponding to the target area.

17. The method according to claim 16, characterized in that, If there are multiple target locations, the scanning guidance information also includes the scanning order of the multiple target locations.

18. An electronic device, characterized in that, include: A transmission component is connected to an ultrasound device so that the electronic device can receive real-time ultrasound images transmitted by the ultrasound device; The display unit includes a display module and a touch module. The display module is used to display real-time ultrasound images, and the touch module is used to receive user input. The processor is electrically connected to the transmission component and the display unit; A memory electrically connected to the processor, the memory storing a computer program executable on the processor, wherein when the processor executes the program, it implements the method of any one of claims 1 to 17.

19. A computer program product, characterized in that, The computer program product includes instructions that, when the instructions are executed, the method of any one of claims 1 to 17 is performed.

Images