An auxiliary decision-making method, system, medium and medical equipment applied to ultrasonic diagnosis and treatment

Abstract

The application is suitable for the medical diagnosis and treatment field, and provides an auxiliary decision-making method and system applied to ultrasonic diagnosis and treatment, a medium and medical equipment.The method comprises the following steps: in response to a selection instruction of a user, a target logical decision workflow selected by the user is loaded; whether a standardized image acquisition workflow is pre-associated with a current decision node of the target logical decision workflow is judged; if yes, the image acquisition workflow associated with the current decision node is activated; a decision option selected by the user for a judgment question of the current decision node is received, and the current decision node, the decision option and an ultrasonic image associated with the decision option are data-associated to form structured decision evidence and are stored; according to a preset logical relationship of the decision option and the target logical decision workflow, a next decision node is determined and entered until a conclusion node is reached.The application solves the problems of insufficient quality, efficiency and reliability of the existing ultrasonic diagnosis and treatment.

Description

Technical Field

[0001] This invention relates to the field of medical diagnosis and treatment technology, and in particular to an auxiliary decision-making method, system, medium and medical device for ultrasound diagnosis and treatment. Background Technology

[0002] Ultrasound diagnosis, due to its advantages of being non-invasive, real-time, portable, and without ionizing radiation, has become an indispensable imaging examination tool in modern clinical medicine. It plays a crucial role, especially in scenarios such as intensive care unit (ICU), emergency department (ER), and point-of-care testing (POCUS). To ensure the comprehensiveness, accuracy, and standardization of diagnosis in complex or emergency situations, a series of standardized diagnostic procedures and protocols have been developed clinically. Examples include the BLUE protocol (Bedside Lung Ultrasound in Emergency) for assessing acute respiratory distress, the RUSH protocol (Rapid Ultrasound in Shock) for investigating the causes of shock, and the procedure for extracorporeal membrane oxygenation (ECMO) in adults.

[0003] However, existing technologies face significant limitations and technical bottlenecks in combining these standardized diagnostic and treatment logics with actual ultrasound and other procedures, mainly in the following aspects:

[0004] First, there is a severe disconnect between diagnostic logic and data acquisition procedures. Currently, the aforementioned diagnostic protocols mainly exist in the form of texts, flowcharts, etc., in textbooks, academic guidelines, or in the doctor's memory. When performing ultrasound examinations, doctors need to rely on their personal memory or consult these external materials during breaks to follow the diagnostic steps. This approach has significant drawbacks: First, it heavily relies on the doctor's personal experience and memory, making it prone to omissions or logical errors for inexperienced doctors or in emergency situations; second, the doctor's cognitive process (following the procedure) is completely separated from the physical operation (ultrasound scanning), making it impossible to form effective real-time guidance and quality control.

[0005] Secondly, there is a lack of real-time decision support linked to imaging data. Although some studies have used models such as decision trees to retrospectively analyze massive amounts of archived clinical data to uncover diagnostic patterns, these approaches remain at the level of "offline big data analysis." In real-time bedside diagnosis, existing ultrasound systems cannot proactively "understand" the diagnostic task the doctor is currently performing, let alone analyze the currently acquired ultrasound images in real time and provide intelligent decision suggestions for the current clinical judgment questions (e.g., "Is there pleural slippage?"). Every judgment made by the doctor relies entirely on their subjective interpretation of dynamic images, which is not only inefficient but also challenges the objectivity and consistency of diagnostic results.

[0006] Secondly, there is the issue of non-standardized image acquisition and the lack of traceability in the diagnostic process. Traditional ultrasound examinations rely heavily on the operator's technique; even for the same patient, images obtained by different doctors at different times can vary significantly. This non-standardization problem is particularly pronounced when precise comparison and quantitative analysis are required. Furthermore, after diagnosis, the final report typically contains only a few static keyframes and a concluding paragraph. The entire logical reasoning process of the diagnosis, the basis for each decision, and the specific scanning context (such as probe position and angle) used to obtain this basis are all lost. This information gap makes case review, teaching and training, and medical quality control extremely difficult.

[0007] In summary, there is an urgent need for a new solution that can deeply integrate standardized diagnostic logic, standardized image acquisition operations, and intelligent real-time data analysis to address the core pain points in current ultrasound diagnosis, such as the disconnect between logic and operation, lack of real-time decision support, and lack of process traceability. Summary of the Invention

[0008] Based on this, the purpose of this invention is to provide an auxiliary decision-making method, system, medium, and medical device for ultrasound diagnosis and treatment, so as to fundamentally solve the problems of insufficient quality, efficiency, and reliability of existing ultrasound diagnosis and treatment.

[0009] According to an embodiment of the present invention, an auxiliary decision-making method for ultrasound diagnosis and treatment includes:

[0010] In response to the user's selection command, the target logical decision-making workflow selected by the user is loaded from multiple stored logical decision-making workflows that are preset based on clinical practice experience or user-defined. The logical decision-making workflow contains multiple decision nodes connected by preset logical relationships, and each decision node corresponds to a judgment question and at least one option for the user to select.

[0011] For the current decision node of the target logic decision workflow, determine whether the current decision node is pre-associated with a standardized image acquisition workflow. The image acquisition workflow includes guidance information for guiding the user to perform standard cross-section scanning. The guidance information includes a human graphical interface indicating the standard cross-section to be scanned, and device parameters matching each standard cross-section.

[0012] If so, the image acquisition workflow associated with the current decision node is activated, and the user is guided to move the ultrasound probe to one or more target standard section positions for standardized ultrasound image acquisition through the human graphical interface in the guidance information. The device parameter status of the target standard section is automatically switched according to the device parameters matching the target standard section in the guidance information.

[0013] Receive the decision option selected by the user for the judgment problem of the current decision node, and associate the current decision node, the decision option, and the ultrasound image associated with the decision option to form structured decision evidence and store it;

[0014] Based on the decision options and the preset logical relationships of the target logical decision workflow, determine and proceed to the next decision node of the target logical decision workflow until the conclusion node of the target logical decision workflow is reached.

[0015] In addition, the auxiliary decision-making method for ultrasound diagnosis and treatment according to the above embodiments of the present invention may also have the following additional technical features:

[0016] Furthermore, the step of activating the image acquisition workflow associated with the current decision node includes:

[0017] On the display interface, pause the user interface of the target logic decision workflow, and load and display the human graphical interface of the image acquisition workflow;

[0018] On the human body graphical interface, the positions of each target standard section that the current decision node is pre-configured in the image acquisition workflow are indicated in turn by highlighting or dynamic icons. The standard section is pre-configured to include standard section name, probe, inspection mode, and quick anomaly annotation.

[0019] For each indicated target standard section position, a preset device parameter state that matches the scanning task of the target standard section is automatically configured. The device parameter state includes at least one of probe mode, imaging frequency, inspection mode and gain.

[0020] After the user completes image acquisition of all target standard cross-sectional positions and manually exits the human body graphical interface, the user interface of the target logic decision-making workflow is restored.

[0021] Furthermore, the step of guiding the user to move the ultrasound probe to one or more target standard section positions for standardized ultrasound image acquisition through the human body graphical interface in the guidance information includes:

[0022] An anatomical model containing the target organ or human body part is presented on the human body graphical interface, and the target standard section position corresponding to the current decision node is marked on the anatomical model;

[0023] Real-time pose data of a positioning sensor integrated with an ultrasonic probe is acquired, the pose data including three-dimensional spatial coordinates and rotation angle;

[0024] The real-time pose data is mapped to a virtual probe on the anatomical model, and real-time navigation feedback is provided based on the deviation between the real-time pose data and the target standard section position to guide the user to adjust the position and orientation of the ultrasound probe.

[0025] When the deviation is less than a preset threshold, it is confirmed that a standard section image has been acquired, and the standard section image is used as the ultrasound image in the structured decision evidence.

[0026] Furthermore, the step of providing real-time navigation feedback includes:

[0027] Calculate the translational and rotational deviations between the real-time pose data of the ultrasonic probe and the position of the target standard section;

[0028] The translational and rotational deviations are converted into graphical instructions, which include at least one of changing the color and transparency of the virtual probe, or displaying directional arrows around the virtual probe.

[0029] The graphical instructions are displayed synchronously on the display interface until both the translation deviation and the rotation deviation are less than their respective preset thresholds.

[0030] Furthermore, the method also includes:

[0031] When the user is acquiring images, the system automatically identifies and analyzes the ultrasound image data acquired in real time by the ultrasound probe to detect specific sonographic indicators corresponding to the judgment problem of the current decision node.

[0032] Based on the detection results, a recommended option for the current decision node is generated and displayed visually in an emphasized manner on the user interface of the target logic decision workflow.

[0033] Furthermore, the step of generating and visually emphasizing a recommended option for the current decision node includes:

[0034] Obtain the image recognition task corresponding to the judgment problem of the current decision node;

[0035] An image algorithm model matching the image recognition task is invoked to extract and analyze features from real-time ultrasound images in order to identify whether there are any preset specific sonographic indicators.

[0036] Based on the analysis output of the image algorithm model, the confidence level of each option is calculated;

[0037] The option with the highest confidence level is selected as the recommended option and displayed in a visually emphasized manner on the user interface.

[0038] Furthermore, the step of associating the current decision node, the decision option, and the ultrasound images associated with the decision option to form structured decision evidence and storing it includes:

[0039] Create a decision evidence object and record the unique identifier of the current decision node and the unique identifier of the decision option selected by the user into the decision evidence object;

[0040] If the image acquisition workflow is activated, the target standard section position information and the storage handle of at least one frame of standardized ultrasound image acquired under guidance are recorded in the decision evidence object.

[0041] If the image acquisition workflow is not activated, the storage handle of the real-time or acquired ultrasound images that the user bases for making decisions is recorded in the decision evidence object.

[0042] Furthermore, the step of determining and entering the next decision node of the target logical decision workflow based on the decision options and the preset logical relationship of the target logical decision workflow includes:

[0043] Access the logical mapping table stored internally in the target logical decision workflow. The logical mapping table maps the unique identifier of each optional decision option of any decision node to the unique identifier of the next decision node.

[0044] Using the unique identifier of the decision option selected by the user for the current decision node as the query key, the logical mapping table is searched to retrieve the unique identifier of the next decision node that is mapped to it.

[0045] Based on the unique identifier of the retrieved next decision node, the next decision node is accurately located and loaded in all decision nodes of the target logical decision workflow, thereby realizing guided non-linear process jump based on the current judgment result.

[0046] Furthermore, the method also includes:

[0047] When the conclusion node of the target logical decision-making workflow is reached, a structured diagnostic report is automatically generated based on all the stored structured decision evidence. The diagnostic report displays the complete execution path of the target logical decision-making workflow in the form of a timeline or tree diagram, and provides interactive links for each decision node on the execution path.

[0048] In response to a user's trigger command for linking any decision node, the system displays the judgment question associated with the triggered decision node, the decision option selected by the user, and the ultrasound image associated with the decision option simultaneously, either side-by-side or in separate areas, on the display interface.

[0049] Another objective of this invention is to provide an auxiliary decision-making system for ultrasound diagnosis and treatment, the system comprising:

[0050] The loading module is used to load the target logical decision workflow selected by the user from multiple preset or user-defined logical decision workflows in response to the user's selection command. The logical decision workflow contains multiple decision nodes connected by preset logical relationships, and each decision node corresponds to a judgment question and at least one option for the user to select.

[0051] The judgment module is used to determine whether the current decision node of the target logic decision workflow is pre-associated with a standardized image acquisition workflow, wherein the image acquisition workflow includes a human graphical interface for guiding users to perform standard cross-sectional scanning.

[0052] The acquisition module is used to activate the image acquisition workflow associated with the current decision node when the judgment module determines that the current decision node is pre-associated with a standardized image acquisition workflow. The module guides the user to move the ultrasound probe to one or more target standard section positions for standardized ultrasound image acquisition through the human graphical interface.

[0053] The storage module is used to receive the decision options selected by the user for the judgment problem of the current decision node, and to associate the current decision node, the decision options, and the ultrasound images associated with the decision options to form structured decision evidence and store it.

[0054] The loop module is used to determine and enter the next decision node of the target logical decision workflow based on the decision options and the preset logical relationship of the target logical decision workflow, until the conclusion node of the target logical decision workflow is reached.

[0055] Another embodiment of the present invention aims to provide a medium storing a program that, when executed by a processor, implements the auxiliary decision-making method for ultrasound diagnosis and treatment as described above.

[0056] Another embodiment of the present invention aims to provide a medical device, including a memory, a processor, and a program stored in the memory and executable on the processor, wherein the processor executes the program to implement the auxiliary decision-making method for ultrasound diagnosis and treatment as described above.

[0057] The auxiliary decision-making method for ultrasound diagnosis and treatment provided in this invention pre-associates decision nodes in the logical decision-making workflow with a standardized image acquisition workflow, and automatically activates the acquisition workflow at specific decision times. This achieves seamless connection and automated guidance from abstract clinical judgment to specific, standardized scanning operations, solving the technical problems of traditional ultrasound diagnosis and treatment, such as doctors relying on memory to execute procedures, high cognitive load, and easy omissions or non-standard operations. This significantly improves the standardization level and execution efficiency of the diagnostic process. By presenting an anatomical model in image acquisition and using positioning sensors to track the ultrasound probe's pose in real time, it provides... Real-time navigation feedback based on spatial deviation calculation enables precise and quantitative guidance during the standard section acquisition process. This solves the problems of difficult section localization and inconsistent image quality caused by insufficient operator experience or anatomical variations in traditional scanning, thus ensuring the quality and repeatability of key diagnostic images. By automatically analyzing the real-time ultrasound image data stream during image acquisition, detecting specific sonographic indicators, and recommending decision options with visual emphasis based on confidence calculation results, this represents a leap from passive tools to active intelligent assistance. It addresses the difficulties, time-consuming nature, and error-proneness of subjective judgment by doctors when faced with complex or blurry images. This provides objective and rapid reference for clinical decision-making, improving diagnostic accuracy. By associating each decision node and user-selected decision option with the ultrasound images used as the basis for judgment, a structured decision evidence is formed and stored, constructing a complete and rigorous chain of decision evidence. This solves the problem of lack of systematic correlation between decisions and image evidence in traditional diagnostic processes, and the difficulty in retrospection, thus greatly enhancing the reliability, transparency, and reviewability of diagnostic conclusions. After the process is completed, a structured diagnostic report is automatically generated, displaying the complete execution path in the form of a timeline or tree diagram and providing interactive links. This achieves comprehensive visualization and traceability of the diagnostic process, solving the limitations of traditional static reports that cannot show diagnostic thinking and are not conducive to review and teaching, thus completely changing the presentation and interaction mode of reports. When guiding standardized scanning, the preset equipment parameter status is automatically configured for each different target section position, realizing intelligent and scenario-based management of equipment status. This solves the problem of users having to frequently manually adjust a large number of complex parameters between different scanning tasks, further simplifying the operation process, shortening the total examination time, and ensuring optimal image quality under various conditions. Therefore, it solves the problems of insufficient quality, efficiency and reliability of existing ultrasound diagnosis and treatment. Attached Figure Description

[0058] Figure 1 This is a flowchart illustrating the auxiliary decision-making method applied to ultrasound diagnosis and treatment according to the first embodiment of the present invention;

[0059] Figure 2 This is a schematic diagram of the auxiliary decision-making system applied to ultrasound diagnosis and treatment according to the second embodiment of the present invention;

[0060] Figure 3 This is a schematic diagram of the structure of the medical device in the third embodiment of the present invention;

[0061] Figure 4 This is a schematic diagram of the logical decision-making workflow in the auxiliary decision-making method applied to ultrasound diagnosis and treatment according to the first embodiment of the present invention;

[0062] Figure 5 This is a schematic diagram of the user interface for loading the target logic decision-making workflow in the medical device display interface of the auxiliary decision-making method applied to ultrasound diagnosis and treatment in the first embodiment of the present invention.

[0063] Figure 6 This is a schematic diagram of the user interface and the human graphical interface simultaneously loading the target logic decision-making workflow in the medical device display interface of the auxiliary decision-making method applied to ultrasound diagnosis and treatment in the first embodiment of the present invention.

[0064] Figure 7 This is a schematic diagram of loading a diagnostic report in the medical device display interface of the auxiliary decision-making method applied to ultrasound diagnosis and treatment in the first embodiment of the present invention.

[0065] The following detailed description of the embodiments will further illustrate the present invention in conjunction with the above-described accompanying drawings. Detailed Implementation

[0066] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0067] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0068] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items. Example 1

[0069] Please see Figure 1 The diagram illustrates an auxiliary decision-making method for ultrasound diagnosis and treatment according to a first embodiment of the present invention. For ease of explanation, only the parts relevant to the embodiments of the present invention are shown. The auxiliary decision-making method for ultrasound diagnosis and treatment provided by the embodiments of the present invention includes:

[0070] Step S10: In response to the user's selection instruction, load the target logical decision workflow selected by the user from multiple stored logical decision workflows that are preset based on clinical practice experience or user-defined. The logical decision workflow contains multiple decision nodes connected by preset logical relationships. Each decision node corresponds to a judgment question and at least one option for the user to select.

[0071] In one embodiment of the present invention, the method is applied to a medical device to provide auxiliary decision-making for users when diagnosing ultrasound images. The processing unit of the medical device, in its initialization or standby state, presents an interactive workflow selection interface on its display screen. This interface displays all available logical decision workflows stored in the workflow template library area of ​​the medical device's memory, in the form of a list, grid, or scrollable menu. These workflows are divided into two main categories: "system presets" (such as international standard processes like "BLUE protocol" and "RUSH protocol") and "user-defined" workflows. Each workflow is accompanied by a brief name and functional description to facilitate user identification and selection. The aforementioned system preset logical decision workflows are derived from the analysis of a large amount of validated clinical case data, combined with authoritative clinical guidelines and expert consensus, to reverse-engineer and summarize optimized decision-making paths based on the clinical manifestations and imaging characteristics of specific diseases. Their design aims primarily to provide clinicians lacking experience in diagnosing specific diseases with a standardized and validated framework for examination and judgment. These pre-set or user-defined diagnostic and treatment processes built into medical devices, based on clinical experience, can effectively assist doctors in conducting systematic examinations, avoiding missed or misdiagnosed cases due to insufficient experience, thereby significantly improving the efficiency and quality of diagnosis and treatment. Especially in areas with relatively weak medical resources, where some departments may lack relevant clinical experience in diagnosis and treatment, or when facing regionally prevalent diseases, this standardized decision support can play a crucial guiding role. Furthermore, embodiments of this invention also support the sharing and deployment of these effective decision-making processes across multiple ultrasound devices through import and export, greatly promoting the dissemination and popularization of high-quality diagnostic and treatment experience. This provides powerful decision support methods for doctors in areas with underdeveloped medical resources or those lacking experience in the diagnosis or treatment of regional diseases, improving the quality and efficiency of diagnosis and treatment.

[0072] Next, when the medical device receives an explicit instruction from the user via a touchscreen or external input device that the user has clicked or selected one of the target logical decision workflows (e.g., the user clicked the "BLUE" workflow), the medical device's workflow engine module is activated and performs a series of loading and parsing actions. Specifically, the workflow engine module first obtains the file handle or database index corresponding to the target logical decision workflow in memory based on the user's selected instruction. This logical decision workflow is typically stored in a structured data format, such as XML (Extensible Markup Language), JSON (JavaScript Object Notation), or a dedicated binary file.

[0073] Then, the workflow engine module reads and parses this structured data file, deserializes its contents, and constructs a dynamic, executable workflow object instance in memory. This parsing process specifically includes: traversing all "decision node" definitions in the file and creating a corresponding node object in memory for each node. Each node object contains at least the following attributes: a unique node ID (e.g., "Node_1"), a string-type judgment question (e.g., "Does pleural spondylolisthesis exist?"), a list containing multiple option objects (e.g., options "exists", "disappears", "optional"), a reference to the associated image acquisition workflow (initially null), and the location of one or more target standard sections corresponding to the image acquisition workflow. Then, the "logical relationships" or "jump rules" sections in the file are parsed. Based on these rules, the workflow engine establishes pointer links between the various node objects in memory. For example, it finds the object with node ID "Node_1" and sets its "yes" option's next pointer to the object instance with node ID "Node_2", and its "no" option's next pointer to another node object instance. In this way, a complete, directed, acyclic graph (or tree) structure is constructed in memory. Simultaneously, during the parsing process, the workflow engine checks if any nodes are associated with image acquisition workflows or other external resources (such as instructional videos, reference images, etc.). If an association is detected, the engine pre-loads the metadata of these resources (such as file paths, protocol names, etc.) into memory and establishes an association with the corresponding node object, enabling rapid invocation when that node is executed subsequently, avoiding the latency caused by real-time lookups.

[0074] Finally, once the entire target logic decision-making workflow is successfully parsed and constructed into a complete object instance in memory, the workflow engine sets the starting node of that instance (usually a pre-marked node or one with the ID "start") as the current decision node. Subsequently, the processing unit extracts the judgment question and option information from this starting node object and renders it to the user interaction area of ​​the display interface, as detailed in [reference needed]. Figure 4 As shown. Specifically, the processing unit displays the "judgment term / conclusion" text of the node in the question area of ​​the interface, and renders each option text in the node's "option list" as an interactive button. At this point, the user-selected target logic decision-making workflow has been successfully loaded and initialized, and the medical device presents the interactive interface of the first decision node, waiting for the user to perform the next operation or judgment. The entire assisted decision-making process officially begins.

[0075] Step S20: For the current decision node of the target logic decision workflow, determine whether the current decision node is pre-associated with a standardized image acquisition workflow;

[0076] In one embodiment of the present invention, if it is determined that the current decision node is pre-associated with a standardized image acquisition workflow, step S30 is executed; otherwise, step S40 is executed.

[0077] Specifically, when the workflow engine moves to a new current decision node based on the decision result of the previous node, the processing unit's decision module is immediately triggered. This decision module first accesses the data object representing the current decision node in memory. As mentioned earlier, this data object is created by parsing a structured data file when the workflow is loaded, and it contains all the node's attributes and related information.

[0078] Next, the decision module specifically queries the predefined attribute fields in the node's data object to identify associated information. These attribute fields include a reference to the associated scanning protocol, which stores the unique identity information of the standardized image acquisition workflow (wiScan) bound to this decision node. This identity information can be descriptive text (e.g., the name of the image acquisition workflow, "Lung Examination Protocol"), a numerical code, or an internal link directly pointing to the corresponding image acquisition workflow data object in memory. It also includes a list of target sections to be scanned, detailing the names of one or more standard body parts or anatomical sections that need to be scanned sequentially in the invoked image acquisition workflow (e.g., "Right First Exploration Point" and "Left First Exploration Point"). The image acquisition workflow contains guidance information to instruct the user on standard section scanning, including a graphical human body interface indicating the required standard sections and the device parameters matching each standard section.

[0079] Then, the judgment module performs a logical judgment on the content of the associated scanning protocol reference attribute field found in the query, checking whether the content of the attribute field is empty or not set. If the content of the field is empty, the output result of the judgment module is "No" (i.e., "Not associated"). This indicates that the execution flow of the current decision node is relatively simple and does not require initiating a complex, guided standardized scanning process. The medical device will then skip the step of activating the image acquisition workflow and directly present the judgment question and options of the node to the user, that is, execute step S40. In one embodiment of the present invention, please refer to the following for details. Figure 5 As shown.

[0080] If the attribute field contains valid identity information (e.g., the text "Lung Examination Protocol"), the judgment module outputs "Yes" (i.e., "Associated"). In addition to returning a positive result, the judgment module also passes this valid identity information (e.g., "Lung Examination Protocol") and the list of target sections to be scanned to the medical device's collaborative controller module. The collaborative controller module uses this identity information as a search keyword to immediately search and match in the medical device's workflow template library area. It precisely locates the definition file of the standardized image acquisition workflow named "Lung Examination Protocol" and begins parsing and preloading it, fully preparing for the next activation operation.

[0081] Therefore, by using this mechanism to check whether the content of specific attribute fields in the node data object is empty, medical devices can quickly and accurately determine whether each decision node needs to trigger a deep, guided image acquisition process, thereby realizing intelligent switching between two decision paths of different complexity.

[0082] Step S30: Activate the image acquisition workflow associated with the current decision node. Guide the user to move the ultrasound probe to one or more target standard section positions for standardized ultrasound image acquisition through the human graphical interface in the guidance information. Automatically switch the device parameter status of the target standard section according to the device parameters matching the target standard section in the guidance information.

[0083] In one embodiment of the present invention, the step of activating the image acquisition workflow associated with the current decision node includes:

[0084] On the display interface, a human-based graphical interface for loading and displaying the image acquisition workflow is provided.

[0085] On the human body graphical interface, the positions of each target standard section in the image acquisition workflow are indicated in sequence by highlighting or dynamic icons. The standard section presets include standard section name, probe, inspection mode, and quick anomaly annotation.

[0086] For each indicated target standard section position, the system automatically configures preset equipment parameter states that match the scanning task of the target standard section. The equipment parameter states include at least one of the following: probe mode, imaging frequency, examination mode, and gain. The examination mode includes, for example, high-penetration heart mode, adult heart mode, or adult lung examination mode, to achieve a deeper level of scenario-based automatic configuration.

[0087] After the user completes image acquisition of all target standard cross-sectional positions and manually exits the human graphical interface, the user interface of the target logic decision-making workflow is restored.

[0088] Specifically, the collaborative controller module first sends a mode switching command to the graphical user interface manager of the medical device. Preferably, the decision-making process interface is not forcibly paused at this time, but the "user interface of the target logic decision-making workflow" and the "human body graphical interface of the image acquisition workflow" that is about to be loaded can be displayed simultaneously in different areas on the same screen, realizing the parallel operation of the two. At the same time, the current state of the interface, including the questions and options that the user has seen, is completely saved in memory so that it can be seamlessly restored later. Next, the interface manager retrieves the corresponding interface layout file and resources (such as human body model, icons, etc.) from the memory according to the identifier of the image acquisition workflow passed by the collaborative controller. Then, the interface manager renders and displays the "human body graphical interface of the image acquisition workflow" displayed in parallel on the screen, which can be specifically referred to in one embodiment of the present invention. Figure 6 As shown.

[0089] Once the graphical human body interface is displayed, the co-controller begins executing the visually guided logic. It first accesses the data structure of the loaded image acquisition workflow, which contains an ordered list defining all target standard section locations to be acquired in this task. The co-controller then retrieves the information for the first target standard section location from this list (including its precise coordinates and name on the anatomical model). The co-controller instructs the interface manager to generate a highlighted, possibly semi-transparent, colored area, or a dynamic icon with a breathing-like flashing effect, at the corresponding location on the anatomical model in the graphical human body interface, clearly marking the section location to be scanned. This strong visual stimulation ensures that the user can accurately identify the first target. After the first section is acquired, clicking "Next" or manually selecting another standard section will automatically turn off the highlighted or dynamic icon and appear at the second target standard section location, and so on, until all sections have been indicated.

[0090] Simultaneously, while indicating each new target standard section position, the co-controller automatically configures the device parameters in parallel. This means it automatically switches the device parameter state of the target standard section based on the matching parameters in the guidance information, ensuring the ultrasound host is in optimal imaging condition and reducing user intervention. The co-controller reads the "preset device parameter state" data packet, precisely bound to the currently indicated "target standard section position," from the image acquisition workflow data structure. This data packet details the parameter values ​​optimized for that section, such as: {"probe mode": "convex array", "imaging frequency": "3.5MHz", "examination mode": "high penetration cardiac mode", "gain": "85%", "depth": "15cm"}. The co-controller then calls the medical device's underlying device control interface, sending this series of parameter values ​​as instructions to the corresponding hardware control unit. For example, it commands the probe switcher to select a convex array probe and the examination mode to select high penetration cardiac mode, commands the signal processor to set the center frequency to 3.5MHz, and commands the gain controller to adjust the total gain to 85%. This configuration process is completely automatic, requiring no manual parameter adjustments from the user, thus ensuring standardized and efficient imaging.

[0091] At this point, the collaborative controller continuously monitors the execution status of the image acquisition workflow. When it detects that all target standard cross-sectional positions in the list have been successfully acquired (usually confirmed by the user), and the user manually exits the human graphical interface, the collaborative controller sends an "exit acquisition mode" command to the interface manager. The interface manager then unloads the current human graphical interface and cleans up its associated resources. Immediately afterwards, the interface manager restores the screen to its original state, that is, restores the previous display state of the user interface for the target logic decision-making workflow, i.e., referring to... Figure 5 As shown. At this point, the user can seamlessly continue their previous thought process, evaluate the information obtained through the standardized scan, and select the appropriate decision options.

[0092] Furthermore, in one embodiment of the present invention, the step of guiding the user through a graphical human body interface to move the ultrasound probe to one or more target standard section positions for standardized ultrasound image acquisition includes:

[0093] An anatomical model containing the target organ or human body part is presented on the human body graphical interface, and the target standard section position corresponding to the current decision node is marked on the anatomical model.

[0094] Acquire real-time pose data from a positioning sensor integrated with an ultrasonic probe. The pose data includes three-dimensional spatial coordinates and rotation angles.

[0095] The real-time pose data is mapped to a virtual probe on the anatomical model, and real-time navigation feedback is provided based on the deviation between the real-time pose data and the target standard section position to guide the user to adjust the position and orientation of the ultrasound probe.

[0096] When the deviation is less than the preset threshold, it is confirmed that a standard section image has been acquired, and the standard section image is used as the ultrasound image in the structured decision evidence.

[0097] Specifically, when the collaborative controller module loads and displays the human graphical interface, it first retrieves a high-precision anatomical model corresponding to the current scanning task (e.g., "inferior vena cava scan") from the medical device's model library. This model is not merely a static image, but an interactive 3D object that can be rotated, scaled, and has its transparency adjusted. It accurately displays the relative positions of the target organ (inferior vena cava) and surrounding anatomical landmarks such as bones and muscles. It should be noted that, considering the differences in hardware configuration and performance among different medical devices, a high-precision 3D model is preferred to provide a stronger sense of space and interactivity; however, under conditions of limited computing resources, a 2D schematic diagram that clearly indicates the anatomical location and probe posture can also be used, achieving the same effective guidance function.

[0098] Next, the co-controller reads the data of the first target standard section position defined in the image acquisition workflow and associated with the current decision node. This data includes not only the precise spatial position of the section in the anatomical model coordinate system (a three-dimensional coordinate point), but also its standard scanning pose (a normal vector defined by three rotation angles). Then, the medical device clearly marks this target standard section position on the anatomical model in the form of a semi-transparent virtual plane or phantom with contour lighting effects, allowing the user to intuitively see the final scanning pose and position that needs to be achieved.

[0099] To achieve synchronization between the physical and virtual worlds, this medical device relies on a high-precision positioning sensor tightly integrated with the ultrasound probe. This sensor can employ electromagnetic tracking technology (operating within the range of a magnetic field generator) or optical tracking technology (requiring capture by an external camera). The device's positioning data processing unit continuously receives raw signals from this sensor at an extremely high frequency (e.g., 60 times per second or higher). This unit then decodes and processes the raw signals, converting them into a set of standardized real-time pose data. This data precisely describes the ultrasound probe's six degrees of freedom in a predetermined three-dimensional coordinate system: three-dimensional spatial coordinates (x, y, z), used to determine the probe's position; and three-dimensional rotation angles (typically pitch, yaw, and roll angles), used to determine the probe's attitude. Upon receiving this series of real-time pose data, the collaborative controller module executes the following core navigation algorithm:

[0100] Virtual probe mapping: The medical device creates a virtual probe model on the anatomical model that matches the appearance of the physical probe. The co-controller applies each set of real-time pose data received to this virtual probe in real time through a pre-calibrated coordinate transformation matrix. In other words, when the user moves or rotates the physical probe in reality, the virtual probe on the screen will also make completely consistent and lag-free synchronous movements.

[0101] Deviation Calculation: After each virtual probe position update, the collaborative controller immediately calculates the spatial deviation between the current pose of the virtual probe and the preset pose of the target standard cross-section position. This calculation is decomposed into two parts: one is the translational distance deviation between the two pose center points, and the other is the angular deviation between the two pose normal vectors.

[0102] Provides real-time navigation feedback: The calculated deviation value is immediately used to drive a series of real-time navigation feedback mechanisms to guide the user to adjust their actions in an intuitive and easy-to-understand way. This feedback can be multimodal, for example, by changing the color of the virtual probe (red when the deviation is large, yellow when it is close, and green when it is in place), adjusting its transparency, or displaying directional and magnitude arrows around it.

[0103] The collaborative controller module continuously monitors the pose deviation value and uses it as the basis for determining whether a standard section has been acquired. When the medical device detects that the translational deviation is less than a preset distance threshold (e.g., 2 mm) and the angular deviation is also less than a preset angle threshold (e.g., 3 degrees), and this "in place" state remains stable for a short period of time (e.g., 200 milliseconds to filter out minor hand tremors), the medical device confirms that the user has accurately aligned the probe with the target standard section. Once confirmed, the collaborative controller immediately sends a "capture" command to the ultrasound host's image acquisition unit. The image acquisition unit then freezes the current ultrasound image frame or acquires a short video clip, saving the data of this frame (or clip) that has been confirmed as the standard section image. Most importantly, when this image is saved, it is automatically associated with its corresponding section name, pose data at the time of acquisition, and other metadata, and is explicitly marked as part of the structured decision evidence for subsequent association, storage, and retrospective use.

[0104] Furthermore, the steps described above for providing real-time navigation feedback include:

[0105] Calculate the translational and rotational deviations between the real-time pose data of the ultrasonic probe and the target standard section position;

[0106] The translation and rotation deviations are converted into graphical instructions, which include at least one of changing the color and transparency of the virtual probe, or displaying directional arrows around the virtual probe.

[0107] The graphical instructions are displayed synchronously on the display interface until the translation deviation and rotation deviation are both less than their respective preset thresholds.

[0108] Specifically, at each medical device refresh cycle (e.g., every 1 / 60th of a second), the co-controller module performs a deviation calculation. This is based on two core input data points: first, the real-time pose data of the ultrasound probe acquired and mapped by the positioning sensor, which can be represented as a data structure containing position vectors (Px, Py, Pz) and pose quaternions or Euler angles (Rx, Ry, Rz); second, the pose data of the target standard section position pre-stored in the image acquisition workflow, which is also a data structure containing target position vectors (Tx, Ty, Tz) and target pose quaternions or Euler angles (TRx, TRy, TRz). The calculation process is precisely decomposed into two independent dimensions:

[0109] Calculating translational deviation: The cooperative controller obtains a scalar value, the translational deviation, by calculating the Euclidean distance between two position vectors. This value intuitively reflects the straight-line distance between the physical probe and the target position in space.

[0110] Calculating rotational deviation: The cooperative controller obtains another scalar value, rotational deviation, by calculating the angle between the two orientations (usually represented by a more robust quaternion). This value reflects the angular difference between the physical probe's orientation and the standard scanning orientation.

[0111] After calculating the translation and rotation deviations, the collaborative controller module immediately converts them into a set of multimodal graphical instructions. This conversion process follows a set of preset mapping rules to convey the correction operation in the most intuitive visual language:

[0112] 1. Changing the color of the virtual probe: The medical device dynamically changes the color of the virtual probe on the screen based on the magnitude of translational and rotational deviations. This color change is tiered; for example, when any deviation value exceeds a large "far-distance threshold," the virtual probe displays red, indicating a large deviation. When the deviation value is between the "far-distance threshold" and the "near-distance threshold," the virtual probe turns yellow, indicating it is approaching. When both deviation values ​​are less than their respective "near-distance thresholds" but still not within the acceptable range, the virtual probe turns blue.

[0113] 2. Adjusting the transparency of the virtual probe: As a complement or alternative to color, the transparency of the virtual probe can also be linked to the deviation value. The greater the deviation, the more transparent the virtual probe becomes or it appears as a wireframe; while the smaller the deviation, the more solid the virtual probe appears.

[0114] 3. Displaying directional guidance arrows: These are the most direct navigation instructions, and the controller further analyzes the direction of the deviation vector. For translational deviations, the medical device displays a 3D arrow around the virtual probe, pointing directly to the target position. The length or size of the arrow is proportional to the translational deviation. The user simply "follows" the arrow to move the physical probe. For rotational deviations, the medical device displays one or more curved rotational arrows on the virtual probe model, clearly indicating how the user should "twist" their wrist to correct the probe's orientation.

[0115] All these converted graphical instructions (color, transparency, arrows) are immediately sent to the graphical user interface manager and synchronously applied and displayed on the virtual probe and its surroundings on the screen in the next frame's display refresh. This "calculation-conversion-display" cycle continues at an extremely high frequency. Therefore, the navigation feedback seen by the user is continuous and real-time. When the user adjusts the physical probe, they will immediately see a decrease in the deviation value and a change in navigation instructions on the screen (e.g., the red arrow shortens, the color changes from red to yellow). This process continues until the co-controller detects that the calculated translational and rotational deviations are both less than their respective preset final thresholds for confirming positioning (e.g., translational deviation < 2mm and rotational deviation < 3 degrees). Once this condition is met, all navigation feedback graphical instructions (such as arrows, warning colors, etc.) disappear, and the virtual probe may turn a bright green with a locking sound effect, marking the successful completion of the navigation task.

[0116] It should be noted that step S40 is also executed after step S30.

[0117] Step S40: Receive the decision options selected by the user for the judgment problem of the current decision node, and associate the current decision node, decision options, and ultrasound images associated with the decision options to form structured decision evidence and store it.

[0118] In one embodiment of the present invention, the step of associating the current decision node, decision options, and ultrasound images associated with the decision options to form structured decision evidence and storing them includes:

[0119] Create a decision evidence object and record the unique identifier of the current decision node and the unique identifier of the user's selected decision option into the decision evidence object;

[0120] If the image acquisition workflow is activated, the target standard section location information and the storage handle of at least one frame of standardized ultrasound image acquired under guidance are recorded in the decision evidence object.

[0121] If the image acquisition workflow is not activated, the storage handle of the real-time or acquired ultrasound images that the user bases for making decisions is recorded in the decision evidence object.

[0122] Specifically, when the medical device presents the decision problem for the current decision node and the options for the user on the display interface, the workflow engine module of the medical device enters a "waiting for user input" state. The medical device continuously listens for instructions from the user interface through its input manager. This can be the user directly clicking option buttons on the touchscreen, or selecting and submitting using navigation keys or confirmation keys on the physical control panel. Once a valid user instruction is received, the input manager parses it into a specific "target decision option" instruction, which contains a unique identifier for the specific option selected by the user (e.g., the text content of the option or a preset numerical code).

[0123] Guided by the decision-making workflow, image acquisition and related diagnostic and treatment operations are performed. The guidance provided by this invention far surpasses that of traditional teaching software. It is not merely a simple demonstration of steps, but rather internalizes complex diagnostic and treatment logic into each interaction. For example, when performing a treatment procedure involving drug dosage calculation, the decision node may require the user to input the patient's weight and automatically calculate and recommend the dosage based on a built-in pharmacological model. When performing a puncture guidance procedure, the image acquisition workflow not only guides probe positioning but may also overlay a virtual puncture path onto the ultrasound image. This guidance method, which tightly integrates specific operational instructions with logical judgments, is more flexible, real-time, and efficient than simple text or video teaching, truly achieving intelligent assistance during the operation process.

[0124] Upon receiving the user's selected decision options, the medical device simultaneously determines the ultrasound image most relevant and reliable for the decision. This determination process is dynamic and depends on the execution of preceding steps.

[0125] If the image acquisition workflow is activated and completed: In this case, the medical device acquires all standardized ultrasound images acquired and stored during the image acquisition workflow execution via probe navigation. These images are typically named and organized according to standard sections, forming an image evidence set. The medical device acquires a storage handle to this image evidence set or an internal reference pointing to its storage location.

[0126] If the image acquisition workflow is not activated: In this case, the medical device acquires the ultrasound image captured in real-time at the moment the user makes a decision or the most recent frame. This is typically a real-time screenshot, or the medical device may briefly freeze and automatically capture the current ultrasound image before the user confirms the decision. The medical device also acquires the storage handle or internal reference to this image.

[0127] After successfully receiving decision options and identifying associated images, the medical device's data association and storage module begins constructing and persisting structured decision evidence. The medical device creates a new decision evidence object in memory. This object is designed as a composite data structure to logically bind multiple pieces of related information together. At this point, the unique identifier of the current decision node, as well as the unique identifier of the user-selected decision option, are recorded in the decision evidence object. Optionally, metadata such as the timestamp of the decision and the user ID are recorded to enhance traceability.

[0128] Furthermore, the storage handle or internal reference of the identified ultrasound image is recorded in the decision evidence object. If the image is a set of image evidence, the reference to the set is recorded; if it is a single frame image, the reference to that single frame image is recorded. This association does not directly embed the image data into the evidence object, but rather establishes a logical link to optimize storage efficiency and query speed.

[0129] Finally, the data association and storage module writes this complete structured decision evidence object into a pre-defined "decision evidence storage area" (e.g., a dedicated database table or file collection) in the medical device's memory for persistent storage. Each decision evidence object is typically marked with a unique transaction ID or sequence number for quick retrieval later.

[0130] Through the above steps, each user's choice in the decision-making process is transformed into traceable decision evidence with clear context and visual evidence, providing a solid data foundation for subsequent report generation, review, and teaching.

[0131] In one embodiment of the present invention, the method further includes:

[0132] When the user is acquiring images, the system automatically identifies and analyzes the ultrasound image data acquired in real time by the ultrasound probe to detect specific sonographic indicators corresponding to the judgment problem of the current decision node.

[0133] Based on the detection results, a recommended option for the current decision node is generated and displayed visually in a way that emphasizes the user interface of the target logic decision workflow.

[0134] Furthermore, the steps described above for generating and visually emphasizing a recommended option for the current decision node include:

[0135] Obtain the image recognition task corresponding to the judgment problem of the current decision node;

[0136] The image algorithm model that matches the image recognition task is invoked to extract and analyze features from real-time ultrasound images in order to identify whether there are any pre-defined specific sonographic indicators.

[0137] Based on the analysis output of the image algorithm model, the confidence level of each option is calculated;

[0138] The option with the highest confidence level is recommended and displayed visually in the user interface.

[0139] Specifically, when a user places the ultrasound probe on a patient to begin scanning, the probe continuously generates raw acoustic scan line data at an extremely high frame rate (e.g., 30 to 60 frames per second). The ultrasound host's beamformer and image processor convert this data into a series of two-dimensional B-mode ultrasound image frames in real time, forming a continuous ultrasound image data stream. The medical device's analysis module subscribes to or intercepts the data stream in real time through an internal interface. For effective analysis, the analysis module first performs necessary preprocessing on each frame, which may include:

[0140] Region of Interest (ROI) localization: If the problem is to determine a specific organ (e.g., to determine heart function), the medical device may first run a fast object detection algorithm (such as a variant of YOLO or SSD) to automatically identify and crop the image region containing the organ, in order to reduce the computational burden of subsequent analysis and eliminate interference.

[0141] Image enhancement: Applying image processing techniques, such as contrast stretching and speckle noise suppression (e.g., using nonlocal mean filtering), to improve image quality and make subsequent feature extraction more robust.

[0142] After preprocessing, the analysis module dynamically invokes a pre-trained image algorithm model to detect specific sonographic indicators based on the judgment question of the current decision node. During the loading of the logical decision workflow, each decision node requiring intelligent assistance has one or more image algorithm model identifiers pre-associated in its data structure. For example, for the question "Does the pleural slippage sign exist?", the medical device invokes a "lung slippage sign detection model" specifically designed to detect dynamic lung features. This model might be a complex model combining optical flow and temporal convolutional networks. The invoked model begins processing the real-time (or short-term) image data stream. For dynamic indicators (such as pleural slippage or inferior vena cava collapse), the model analyzes pixel displacement and change patterns between consecutive frames. For example, the "lung slippage sign detection model" calculates the motion vector field of pixels near the pleural line and determines whether its motion pattern conforms to the typical characteristics of "ant crawling". For static indicators (such as the presence of "B-lines" or "pulmonary consolidation"), the model might use a deep convolutional neural network (CNN), such as ResNet or EfficientNet, to extract deep texture and structural features from the image and compare them with known B-line or consolidation patterns in the model library. After analysis, the model outputs a quantitative or qualitative detection result. For example, for pleural slippage, the output might be "present," "absent," or a probability value indicating the likelihood of its presence; for B-lines, the output might be the number of B-lines.

[0143] After obtaining the automatic detection results, the analysis module transforms them into a user-friendly recommendation option. The medical device maps the model's detection results to a specific option at the current decision node according to a preset logical rule. For example, if the "lung spondylolisthesis detection model" outputs "not present," the medical device will automatically map this result to the "not present" option on the interface. Once the recommended option is determined, the analysis module generates a "recommendation display" instruction, which contains a unique identifier for the recommended option. This instruction is sent to the graphical user interface manager. Upon receiving it, the interface manager immediately visually emphasizes the recommended option button on the "target logical decision workflow user interface." This emphasis can be varied and eye-catching, including: changing the background color of the option button to a striking color (such as green or blue); adding a glowing or breathing-like flashing border to the option button; and displaying a small "AI Recommendation" or "Smart Tip" icon next to the option. This recommendation is dynamically updated; if the user moves the probe during the scan, causing changes in the real-time image content, the analysis module will re-analyze, and the recommended option may change accordingly.

[0144] Step S50: Based on the decision options and the preset logical relationship of the target logical decision workflow, determine and enter the next decision node of the target logical decision workflow until the conclusion node of the target logical decision workflow is reached.

[0145] In one embodiment of the present invention, the step of determining and entering the next decision node of the target logic decision workflow based on the decision options and the preset logical relationship of the target logic decision workflow includes:

[0146] Access the logical mapping table stored internally in the target logical decision workflow. The logical mapping table maps the unique identifier of each optional decision option of any decision node to the unique identifier of the next decision node.

[0147] The unique identifier of the decision option selected by the user for the current decision node is used as the query key to search in the logical mapping table to retrieve the unique identifier of the next decision node that is mapped to it.

[0148] Based on the unique identifier of the next decision node retrieved, the next decision node is accurately located and loaded in all decision nodes of the target logical decision workflow, thereby realizing guided non-linear process jump based on the current judgment result.

[0149] When the workflow engine module receives explicit information about the user-selected "target decision option" (e.g., the text "Yes") from the input manager, it immediately searches and matches it within the decision node data object of the current active state. The workflow engine accesses a dedicated data structure within the current decision node data object that stores "preset logical relationships." A key feature of this embodiment is the use of a logical mapping-based jump mechanism, which makes the entire decision-making process highly guided. Unlike traditional linear processes, each step here is based on the judgment result of the previous step, pointing to the next most relevant checkpoint or conclusion through a rigorous logical chain, ensuring the coherence and systematic nature of the diagnostic approach. This is the core difference between the guided features of this embodiment and a simple step list. This data structure is typically a mapping table, constructed by parsing preset logical relationship data when the workflow is loaded. Essentially, it is a global, two-dimensional lookup table or hash table, designed to efficiently answer the question, "After selecting an option at a certain node, where should I go next?" Specifically, each row of this mapping table defines a complete jump rule, which uses a composite key—the unique identifier of a decision node and the unique identifier of an optional decision option under that node—to map one-to-one with the unique identifier of the next decision node. The "query item" of the mapping table is the descriptive text or internal code of each optional option under that node, while its "result item" is the identity information (i.e., "target node identity information") of the next decision node to jump to when that option is selected. For example, refer to... Figure 4 As shown, for the first decision node (judging the term "whether pleural spondylolisthesis exists"), its logical relationship mapping table may be as follows: When "exists" is selected, it jumps to the decision node with the identity information "second node". When "optional" is selected, it jumps to the decision node with the identity information "third node". When "disappears" is selected, it jumps to the decision node with the identity information "fourth node". It should also be noted that, referring to... Figure 5 As shown, the first decision node also has a "previous step" decision option. That is, when medical staff determine that the previous decision option is wrong or need to reselect the target logic decision workflow, they can also choose the "previous step" decision option to roll back to the previous decision node or the logic decision workflow to be selected.

[0150] Next, the workflow engine uses the unique identifier of the user-selected decision option ("exists") together with the unique identifier of the current decision node (the first decision node) to form a precise query key, which is used as the query item. The engine searches in this mapping table and successfully matches the corresponding result item, which is the identity information of the next target node ("the second node").

[0151] After successfully identifying the identity information of the next target node ("second node"), the workflow engine begins the node switching and loading operation. Using the unique identifier of the next decision node ("second node"), the workflow engine performs a traversal or index search through the set of all decision nodes included in the target logical decision workflow to precisely locate the data object of the matching decision node. An internal "current node pointer" used to mark the current active position within the medical device is updated from pointing to the old node data object (first node) to pointing to the newly found target node data object (second node). This pointer switching signifies that the logical flow has officially entered the next decision node at the program level. Once the active node pointer is updated, the medical device's graphical user interface manager immediately extracts the corresponding judgment question and option list from the new active node data object (second node) and renders it on the display interface for the user. The above "receiving options -> matching logic -> entering a new node" constitutes a complete driving loop. The workflow engine continuously repeats this loop, driving the entire logical decision workflow forward step by step. The loop terminates when the workflow engine, after matching logical relationships, finds that the next target node is marked as a "conclusion node." Conclusion nodes are explicitly marked with "Conclusion" or a similar identifier in the type attribute of their data objects. When the workflow engine is about to enter a new node, it first checks its type. Once it detects that the next node is a conclusion node (for example, the process eventually reaches a node with the identity information "Conclusion Node A"), the workflow engine stops its driving loop. It extracts the final diagnostic conclusion or treatment recommendation (e.g., "pneumothorax") from the conclusion node's data object and presents it in a prominent, summarizing manner on the display interface, marking the official end of the execution process of this target logical decision workflow.

[0152] In one embodiment of the present invention, the method further includes:

[0153] When the conclusion node of the target logical decision-making workflow is reached, a structured diagnostic report is automatically generated based on all the stored structured decision evidence. The diagnostic report shows the complete execution path of the target logical decision-making workflow in the form of a timeline or tree diagram, and provides interactive links for each decision node on the execution path.

[0154] In response to the user's trigger command for any decision node link, the system displays the judgment question associated with the triggered decision node, the decision option selected by the user, and the ultrasound image associated with the decision option simultaneously, either side-by-side or in separate areas, on the display interface.

[0155] Specifically, after the conclusion node is triggered, the medical device's report generation module first retrieves and aggregates all stored "structured decision evidence" objects related to the current diagnostic process from the medical device's "decision evidence storage area" based on the unique session identifier of the currently executing workflow. These evidence objects are arranged in chronological order of their creation, completely recording each step chosen by the user. Upon receiving this ordered list of evidence objects, the report generation module does not generate a traditional, static plain text document. Instead, it automatically generates a structured, interactive diagnostic report. This generation process includes rendering the report in two main visual formats based on user preferences or the medical device's default settings. One is a timeline format, linearly arranging each decision point according to the chronological order of the decisions; the other is a more logical tree diagram format, clearly showing the user's choices at each fork in the road and the unique path leading to the conclusion. When generating the timeline or tree diagram, the report generation module iterates through each decision evidence object. It extracts the text content of "judgment question" and "user's selected decision option" from each object and displays this text as node labels on the graphical interface, thereby showing the complete execution path. In one embodiment of the invention, referring to... Figure 7 As shown. It should be noted that each decision node displayed in the report is programmed to be an interactive hyperlink or a clickable button.

[0156] Once the generated structured diagnostic report is displayed on the interface, users (e.g., senior physicians reviewing the report) can interact with it. When a user clicks on any decision node link in the report's tree diagram or timeline using a mouse or finger, the medical device's user interface manager captures this trigger command. This command includes a unique identifier for the decision node corresponding to the clicked link. Upon receiving this command, the report interaction module uses this identifier to immediately and precisely retrieve the "structured decision evidence" object associated with the triggered decision node from the previously aggregated list of evidence objects. After retrieving the corresponding evidence object, the medical device performs a multi-dimensional information synchronization display operation. It dynamically divides the display interface into several side-by-side or segmented display panels. For example, in the first area (e.g., the upper left panel), it clearly displays the judgment question associated with the triggered decision node extracted from the evidence object. In the second area (e.g., the lower left panel), it clearly displays the decision option selected by the user at that time. In the third and most important area (e.g., the main right panel), the medical device displays the ultrasound image associated with that decision option. It checks whether the associated images in the evidence object are a single image or a set of image evidence. If the latter, it presents them as a list of cross-sections with thumbnails, allowing users to click to view larger images and dynamic images of each standard cross-section. In this way, the diagnostic report is no longer just a static statement of conclusion, but a dynamic, explorable, evidence-driven record of the entire diagnostic process. Users can freely and non-linearly review any stage of the entire diagnostic process and simultaneously compare judgments, selections, and image evidence, greatly enhancing the report's transparency, credibility, and educational value.

[0157] In summary, the auxiliary decision-making method for ultrasound diagnosis and treatment described in the above embodiments of the present invention, by pre-associating decision nodes in the logical decision-making workflow with a standardized image acquisition workflow and automatically activating the acquisition workflow at specific decision times, achieves seamless connection and automated guidance from abstract clinical judgment to specific, standardized scanning operations. This solves the technical problems of traditional ultrasound diagnosis and treatment, such as doctors relying on memory to execute procedures, high cognitive load, and the tendency for operational omissions or non-standard procedures to occur, thereby significantly improving the standardization level and execution efficiency of the diagnostic process. Furthermore, by presenting an anatomical model in image acquisition and utilizing positioning sensors to track the ultrasound probe position in real time, the method significantly improves the standardization level and execution efficiency of the diagnostic process. The system provides real-time navigation feedback based on spatial deviation calculations, enabling precise and quantitative guidance during the acquisition of standard sections. This solves the problems of difficult section positioning and inconsistent image quality caused by insufficient operator experience or anatomical variations in traditional scanning, thus ensuring the quality and repeatability of key diagnostic images. By automatically analyzing real-time ultrasound image data streams during image acquisition, detecting specific sonographic indicators, and recommending decision options based on confidence calculations using visual emphasis, the system represents a leap from passive tools to proactive intelligent assistance. This addresses the difficulties, time-consuming nature, and error-proneness of subjective judgment for doctors when facing complex or blurry images. This addresses the problem of lacking a systematic connection between decisions and imaging evidence in traditional diagnostic processes, thus improving diagnostic accuracy. By linking each decision node and user-selected options with the ultrasound images used as the basis for judgment, structured decision evidence is formed and stored, constructing a complete and rigorous chain of decision evidence. This solves the problem of a lack of systematic connection between decisions and imaging evidence in traditional diagnostic processes, and the difficulty in retrospective analysis, thereby greatly enhancing the reliability, transparency, and auditability of diagnostic conclusions. Furthermore, after the process is completed, a structured data structure is automatically generated, displaying the complete execution path in timeline or tree diagram format and providing interactive links. The diagnostic report achieves comprehensive visualization and traceability of the diagnostic process, overcoming the limitations of traditional static reports that cannot demonstrate diagnostic thinking and are unsuitable for review and teaching. This fundamentally changes the presentation and interaction mode of reports. By automatically configuring pre-set equipment parameters for each different target section location during guided standardized scanning, it achieves intelligent and scenario-based management of equipment status. This eliminates the tedious manual adjustment of numerous complex parameters between different scanning tasks, further simplifying the operation process, shortening the total examination time, and ensuring optimal image quality under various conditions. Therefore, it addresses the shortcomings in quality, efficiency, and reliability of existing ultrasound diagnostic and treatment methods. Example 2

[0158] Please see Figure 2This is a schematic diagram of an auxiliary decision-making system for ultrasound diagnosis and treatment provided in the second embodiment of the present invention. For ease of explanation, only the parts related to the embodiment of the present invention are shown. The system includes:

[0159] The loading module 11 is used to load the target logical decision workflow selected by the user from a plurality of logical decision workflows that are preset based on clinical practice experience or user-defined in response to the user's selection command. The logical decision workflow includes a plurality of decision nodes connected by preset logical relationships, and each decision node corresponds to a judgment question and at least one option for the user to select.

[0160] The judgment module 12 is used to determine whether the current decision node of the target logic decision workflow is pre-associated with a standardized image acquisition workflow. The image acquisition workflow includes guidance information for guiding the user to perform standard section scanning. The guidance information includes a human graphical interface indicating the standard section to be scanned, and device parameters matching each standard section.

[0161] The acquisition module 13 is used to activate the image acquisition workflow associated with the current decision node when the judgment module determines that the current decision node is pre-associated with a standardized image acquisition workflow. The module guides the user to move the ultrasound probe to one or more target standard sections for standardized ultrasound image acquisition through the human graphical interface in the guidance information. The module also automatically switches the device parameter status of the target standard section according to the device parameters matching the target standard section in the guidance information.

[0162] Storage module 14 is used to receive the decision option selected by the user for the judgment problem of the current decision node, and to associate the current decision node, the decision option, and the ultrasound image associated with the decision option to form structured decision evidence and store it.

[0163] The loop module 15 is used to determine and enter the next decision node of the target logic decision workflow based on the decision options and the preset logical relationship of the target logic decision workflow, until the conclusion node of the target logic decision workflow is reached.

[0164] Furthermore, in one embodiment of the present invention, the acquisition module 13 includes:

[0165] The first loading unit is used to load and display the human graphical interface of the image acquisition workflow on the display interface.

[0166] The indicator unit is used to sequentially indicate the positions of each target standard section that the current decision node is pre-configured in the image acquisition workflow on the human graphical interface by highlighting or dynamic icons. The standard section is pre-configured to include standard section name, probe, inspection mode, and quick anomaly annotation.

[0167] The parameter configuration unit is used to automatically configure a preset device parameter state that matches the scanning task of the target standard section for each indicated target standard section position. The device parameter state includes at least one of probe mode, imaging frequency, inspection mode and gain.

[0168] The second loading unit is used to restore the user interface of the target logic decision-making workflow after the user has completed image acquisition of all target standard cross-sectional positions and manually exited the human body graphical interface.

[0169] Furthermore, in one embodiment of the present invention, the acquisition module 13 includes:

[0170] The display unit is used to present an anatomical model containing a target organ or human body part on the human graphical interface, and to mark the target standard section position corresponding to the current decision node on the anatomical model.

[0171] The data acquisition unit is used to acquire real-time pose data of the positioning sensor integrated with the ultrasonic probe, the pose data including three-dimensional spatial coordinates and rotation angle;

[0172] The navigation feedback unit is used to map the real-time pose data to a virtual probe on the anatomical model, and to provide real-time navigation feedback based on the deviation between the real-time pose data and the target standard section position, so as to guide the user to adjust the position and orientation of the ultrasound probe.

[0173] An ultrasound image acquisition unit is used to confirm that a standard section image has been acquired when the deviation is less than a preset threshold, and to use the standard section image as ultrasound image in structured decision evidence.

[0174] Furthermore, in one embodiment of the present invention, the navigation feedback unit is used for:

[0175] Calculate the translational and rotational deviations between the real-time pose data of the ultrasonic probe and the position of the target standard section;

[0176] The translational and rotational deviations are converted into graphical instructions, which include at least one of changing the color and transparency of the virtual probe, or displaying directional arrows around the virtual probe.

[0177] The graphical instructions are displayed synchronously on the display interface until both the translation deviation and the rotation deviation are less than their respective preset thresholds.

[0178] Furthermore, in one embodiment of the present invention, the system further includes:

[0179] The analysis module is used to automatically identify and analyze the ultrasound image data acquired in real time by the ultrasound probe when the user is acquiring images, so as to detect specific sonographic indicators corresponding to the judgment problem of the current decision node.

[0180] The recommendation module is used to generate and visually emphasize a recommended option for the current decision node on the user interface of the target logical decision workflow based on the detection results.

[0181] Furthermore, in one embodiment of the present invention, the recommendation module includes:

[0182] The task acquisition unit is used to acquire the image recognition task corresponding to the judgment problem of the current decision node;

[0183] The feature analysis unit is used to call the image algorithm model that matches the image recognition task to extract and analyze features from real-time ultrasound images in order to identify whether there are preset specific sonographic indicators.

[0184] The confidence calculation unit is used to calculate the confidence level of each option based on the analysis output of the image algorithm model.

[0185] The recommendation unit is used to select the option with the highest confidence level as the recommended option and display it on the user interface with visual emphasis.

[0186] Furthermore, in one embodiment of the present invention, the storage module 14 includes:

[0187] An object creation unit is used to create a decision evidence object and record the unique identifier of the current decision node and the unique identifier of the decision option selected by the user into the decision evidence object;

[0188] The first storage unit is used to record the target standard section position information and the storage handle of at least one frame of standardized ultrasound image acquired under guidance into the decision evidence object if the image acquisition workflow is activated.

[0189] The second storage unit is used to record the storage handle of the real-time or acquired ultrasound images that the user bases for making decisions into the decision evidence object if the image acquisition workflow is not activated.

[0190] Furthermore, in one embodiment of the present invention, the loop module 15 includes:

[0191] The access unit is used to access the logical mapping table stored internally in the target logical decision workflow. The logical mapping table maps the unique identifier of each optional decision option of any decision node to the unique identifier of the next decision node.

[0192] The lookup unit is used to search the logical mapping table using the unique identifier of the decision option selected by the user for the current decision node as the query key, so as to retrieve the unique identifier of the next decision node mapped to it.

[0193] The loading unit is used to accurately locate and load the next decision node in all decision nodes of the target logical decision workflow based on the unique identifier of the retrieved next decision node, thereby realizing guided non-linear process jump based on the current judgment result.

[0194] Furthermore, in one embodiment of the present invention, the system further includes:

[0195] The diagnostic report generation module is used to automatically generate a structured diagnostic report based on all stored structured decision evidence when the conclusion node of the target logical decision workflow is reached. The diagnostic report displays the complete execution path of the target logical decision workflow in the form of a timeline or tree diagram, and provides interactive links for each decision node on the execution path.

[0196] The display module is used to respond to the user's trigger command for linking any decision node, and to simultaneously display the judgment question associated with the triggered decision node, the decision option selected by the user, and the ultrasound image associated with the decision option on the display interface, either side-by-side or in separate areas.

[0197] The auxiliary decision-making system for ultrasound diagnosis and treatment provided in this embodiment of the invention has the same implementation principle and technical effects as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment. Example 3

[0198] In another aspect, the present invention also proposes a medical device, please refer to [link / reference needed]. Figure 3 The image shows a medical device according to a third embodiment of the present invention, including a memory 20, a processor 10, and a program 30 stored in the memory 20 and executable on the processor. When the processor 10 executes the program 30, it implements the auxiliary decision-making method for ultrasound diagnosis and treatment as described in the above embodiment.

[0199] In some embodiments, the processor 10 may be a central processing unit (CPU), controller, microcontroller, microprocessor or other data processing chip, used to run program code stored in memory 20 or process data, such as executing access restriction programs.

[0200] The memory 20 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (such as SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 20 can be an internal storage unit of the medical device, such as the hard disk of the medical device. In other embodiments, the memory 20 can also be an external storage device of the medical device, such as a SmartMedia Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the medical device. Furthermore, the memory 20 can include both internal and external storage units of the medical device. The memory 20 can be used not only to store application software and various types of data installed on the medical device, but also to temporarily store data that has been output or will be output.

[0201] It should be pointed out that, Figure 3 The structure shown does not constitute a limitation on the medical device. In other embodiments, the medical device may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0202] This invention also provides a medium storing a program that, when executed by a processor, implements the auxiliary decision-making method for ultrasound diagnosis and treatment as described in the above embodiments.

[0203] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units or modules as needed, that is, the internal structure of the storage device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application.

[0204] Those skilled in the art will understand that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "medium" can mean any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0205] More specific examples of media (a non-exhaustive list) include: electrical connections (electronic devices) with one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Additionally, the media can even be paper or other suitable media on which the program can be printed, as the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in memory.

[0206] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0207] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0208] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A decision support method for ultrasound diagnosis and treatment, characterized in that, The method includes: In response to the user's selection command, the target logical decision-making workflow selected by the user is loaded from multiple stored logical decision-making workflows that are preset based on clinical practice experience or user-defined. The logical decision-making workflow contains multiple decision nodes connected by preset logical relationships, and each decision node corresponds to a judgment question and at least one option for the user to select. For the current decision node of the target logic decision workflow, determine whether the current decision node is pre-associated with a standardized image acquisition workflow. The image acquisition workflow includes guidance information for guiding the user to perform standard cross-section scanning. The guidance information includes a human graphical interface indicating the standard cross-section to be scanned, and device parameters matching each standard cross-section. If so, the image acquisition workflow associated with the current decision node is activated, and the user is guided to move the ultrasound probe to one or more target standard section positions for standardized ultrasound image acquisition through the human graphical interface in the guidance information. The device parameter status of the target standard section is automatically switched according to the device parameters matching the target standard section in the guidance information. Receive the decision option selected by the user for the judgment problem of the current decision node, and associate the current decision node, the decision option, and the ultrasound image associated with the decision option to form structured decision evidence and store it; Based on the decision options and the preset logical relationship of the target logical decision workflow, determine and enter the next decision node of the target logical decision workflow until the conclusion node of the target logical decision workflow is reached; The method further includes: When the conclusion node of the target logical decision-making workflow is reached, a structured diagnostic report is automatically generated based on all the stored structured decision evidence. The diagnostic report displays the complete execution path of the target logical decision-making workflow in the form of a timeline or tree diagram, and provides interactive links for each decision node on the execution path. In response to a user's trigger command for linking any decision node, the system displays the judgment question associated with the triggered decision node, the decision option selected by the user, and the ultrasound image associated with the decision option simultaneously, either side-by-side or in separate areas, on the display interface.

2. The auxiliary decision-making method for ultrasound diagnosis and treatment according to claim 1, characterized in that, The step of activating the image acquisition workflow associated with the current decision node includes: On the display interface, the human graphical interface of the image acquisition workflow is loaded and displayed; On the human body graphical interface, the positions of each target standard section that the current decision node is pre-configured in the image acquisition workflow are indicated in turn by highlighting or dynamic icons. The standard section is pre-configured to include standard section name, probe, inspection mode, and quick anomaly annotation. For each indicated target standard section position, a preset device parameter state that matches the scanning task of the target standard section is automatically configured. The device parameter state includes at least one of probe mode, imaging frequency, inspection mode and gain. After the user completes image acquisition of all target standard cross-sectional positions and manually exits the human body graphical interface, the user interface of the target logic decision-making workflow is restored.

3. The auxiliary decision-making method for ultrasound diagnosis and treatment according to claim 1 or 2, characterized in that, The step of guiding the user through the graphical human interface in the guidance information to move the ultrasound probe to one or more target standard section positions for standardized ultrasound image acquisition includes: An anatomical model containing the target organ or human body part is presented on the human body graphical interface, and the target standard section position corresponding to the current decision node is marked on the anatomical model; Real-time pose data of a positioning sensor integrated with an ultrasonic probe is acquired, the pose data including three-dimensional spatial coordinates and rotation angle; The real-time pose data is mapped to a virtual probe on the anatomical model, and real-time navigation feedback is provided based on the deviation between the real-time pose data and the target standard section position to guide the user to adjust the position and orientation of the ultrasound probe. When the deviation is less than a preset threshold, it is confirmed that a standard section image has been acquired, and the standard section image is used as the ultrasound image in the structured decision evidence.

4. The auxiliary decision-making method for ultrasound diagnosis and treatment according to claim 3, characterized in that, The steps for providing real-time navigation feedback include: Calculate the translational and rotational deviations between the real-time pose data of the ultrasonic probe and the position of the target standard section; The translational and rotational deviations are converted into graphical instructions, which include at least one of changing the color and transparency of the virtual probe, or displaying directional arrows around the virtual probe. The graphical instructions are displayed synchronously on the display interface until both the translation deviation and the rotation deviation are less than their respective preset thresholds.

5. The auxiliary decision-making method for ultrasound diagnosis and treatment according to claim 1, characterized in that, The method further includes: When the user is acquiring images, the system automatically identifies and analyzes the ultrasound image data acquired in real time by the ultrasound probe to detect specific sonographic indicators corresponding to the judgment problem of the current decision node. Based on the detection results, a recommended option for the current decision node is generated and displayed visually in an emphasized manner on the user interface of the target logic decision workflow.

6. The auxiliary decision-making method for ultrasound diagnosis and treatment according to claim 5, characterized in that, The step of generating and visually emphasizing a recommended option for the current decision node includes: Obtain the image recognition task corresponding to the judgment problem of the current decision node; An image algorithm model matching the image recognition task is invoked to extract and analyze features from real-time ultrasound images in order to identify whether there are any preset specific sonographic indicators. Based on the analysis output of the image algorithm model, the confidence level of each option is calculated; The option with the highest confidence level is selected as the recommended option and displayed in a visually emphasized manner on the user interface.

7. The auxiliary decision-making method for ultrasound diagnosis and treatment according to claim 1, characterized in that, The step of associating the current decision node, the decision option, and the ultrasound images associated with the decision option to form structured decision evidence and storing it includes: Create a decision evidence object and record the unique identifier of the current decision node and the unique identifier of the decision option selected by the user into the decision evidence object; If the image acquisition workflow is activated, the target standard section position information and the storage handle of at least one frame of standardized ultrasound image acquired under guidance are recorded in the decision evidence object. If the image acquisition workflow is not activated, the storage handle of the real-time or acquired ultrasound images that the user bases for making decisions is recorded in the decision evidence object.

8. The auxiliary decision-making method for ultrasound diagnosis and treatment according to claim 1, characterized in that, The step of determining and proceeding to the next decision node in the target logic decision workflow based on the decision options and the preset logical relationship of the target logic decision workflow includes: Access the logical mapping table stored internally in the target logical decision workflow. The logical mapping table maps the unique identifier of each optional decision option of any decision node to the unique identifier of the next decision node. Using the unique identifier of the decision option selected by the user for the current decision node as the query key, the logical mapping table is searched to retrieve the unique identifier of the next decision node that is mapped to it. Based on the unique identifier of the retrieved next decision node, the next decision node is accurately located and loaded in all decision nodes of the target logical decision workflow, thereby realizing guided non-linear process jump based on the current judgment result.

9. A decision support system for ultrasound diagnosis and treatment, characterized in that, The system includes: The loading module is used to load the target logical decision workflow selected by the user from multiple preset or user-defined logical decision workflows in response to the user's selection command. The logical decision workflow contains multiple decision nodes connected by preset logical relationships, and each decision node corresponds to a judgment question and at least one option for the user to select. The judgment module is used to determine whether the current decision node of the target logic decision workflow is pre-associated with a standardized image acquisition workflow, wherein the image acquisition workflow includes a human graphical interface for guiding users to perform standard cross-sectional scanning. The acquisition module is used to activate the image acquisition workflow associated with the current decision node when the judgment module determines that the current decision node is pre-associated with a standardized image acquisition workflow. The module guides the user to move the ultrasound probe to one or more target standard section positions for standardized ultrasound image acquisition through the human graphical interface. The storage module is used to receive the decision options selected by the user for the judgment problem of the current decision node, and to associate the current decision node, the decision options, and the ultrasound images associated with the decision options to form structured decision evidence and store it. The loop module is used to determine and enter the next decision node of the target logical decision workflow based on the decision options and the preset logical relationship of the target logical decision workflow, until the conclusion node of the target logical decision workflow is reached. The system also includes: The diagnostic report generation module is used to automatically generate a structured diagnostic report based on all stored structured decision evidence when the conclusion node of the target logical decision workflow is reached. The diagnostic report displays the complete execution path of the target logical decision workflow in the form of a timeline or tree diagram, and provides interactive links for each decision node on the execution path. The display module is used to respond to the user's trigger command for linking any decision node, and to simultaneously display the judgment question associated with the triggered decision node, the decision option selected by the user, and the ultrasound image associated with the decision option on the display interface, either side-by-side or in separate areas.

10. A medium storing a program, characterized in that, When the program is executed by the processor, it implements the auxiliary decision-making method for ultrasound diagnosis and treatment as described in any one of claims 1-8.

11. A medical device, characterized in that, It includes a memory, a processor, and a program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the auxiliary decision-making method for ultrasound diagnosis and treatment as described in any one of claims 1-8.

Images