An ACP-based data analysis system and method for guiding scenario interaction

Abstract

The application discloses a data analysis system and method based on ACP guiding scene interaction, and relates to the technical field of data analysis.The system comprises a portrait construction module, a scene guiding module, a selection decision module and a transmission feedback module.The technical key points are as follows: a user portrait is configured, a rule matching is introduced, and an immersive animation scene suitable for the user portrait is generated;through executing scene guiding, a comprehensive guiding index is obtained, a fluctuation trend is identified, and a scene advancing rate is dynamically adjusted; a wish card suitable for the current immersive animation scene is loaded, a drag selection behavior and an operation time sequence are identified, a target data packet is generated by combining a pre-constructed decision coordinate system to execute multi-terminal data alignment, variance under the same decision node is calculated, and a personal ACP preference report containing consensus items and high-attention conflict items is generated; and the application improves the accuracy of ACP guiding and expression.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, specifically to a data analysis system and method based on ACP-guided scenario interaction. Background Technology

[0002] As my country's aging population accelerates, the public's demand for dignified and high-quality end-of-life care is becoming increasingly urgent. ACP (Advanced Care Process) is a continuous process that supports individuals in clarifying their future medical care goals and preferences based on their own values ​​while they are of sound mind, and communicating with staff and family members. It is a core means of protecting patient autonomy and improving the quality of medical decision-making, and its promotion and practice are of great significance.

[0003] Traditional data analysis techniques based on ACP-guided scenario interaction suffer from the following technical shortcomings: Firstly, traditional techniques often employ questionnaires or rigid clinical interview models. While digital tools have emerged with technological advancements, these tools are often limited and fail to accurately capture user identity and clinical urgency. They cannot adaptively adjust the content hierarchy of the guidance scenario based on user characteristics or dynamically adjust the guidance scenario based on user physiological indicators or interactive behaviors. For example, directly pushing highly stimulating medical decision-making content (such as intubation or resuscitation) to highly sensitive individuals can easily trigger psychological stress and resistance, leading to interruptions in the guidance process and reducing promotion efficiency and adaptability. Secondly, existing systems typically only support independent operation by a single account, lacking functional modules that can link patient and family data and enable real-time collaboration. Furthermore, they cannot capture the frequency of user touch withdrawal during interaction. For the same decision point, patients and families cannot quantify their respective intentions or provide visual conflict warnings, which not only reduces user experience but also affects the authenticity and accuracy of intention data. Summary of the Invention

[0004] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a data analysis system and method based on ACP-guided scenario interaction. Based on multi-dimensional data, it generates an initial immersive animation scenario adapted to the user profile through three rule matching steps. During scenario-guided interaction, it calculates a comprehensive guidance index in real time and identifies fluctuation trends to dynamically adjust the scenario progression rate and content level. It loads wish cards adapted to the current scenario level, calculates the final weight based on drag-and-drop coordinates and operation sequence, and generates a target data package. Through multi-terminal data alignment and variance calculation, it quantifies the dispersion of intentions at the same decision node, generating a personal ACP preference report containing consensus items and highly conflicting items. This improves the accuracy of intention data identification and solves the problems mentioned in the background technology.

[0005] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: In the first aspect, this application provides a data analysis system based on ACP-guided scenario interaction, the system comprising: a profile building module: used to collect multidimensional data of users, configure user profiles, introduce rule matching, and generate immersive animation scenarios adapted to user profiles; Contextual Guidance Module: Uses immersive animated contexts for interaction, and acquires touch trajectories and physiological indicators during the interaction to perform contextual guidance and obtain a comprehensive guidance index; Based on the comprehensive guidance index, it identifies fluctuation trends and dynamically adjusts the contextual progression rate; Selection Decision Module: This module is used to execute selection decision strategies based on a comprehensive guidance index, load wish cards that are compatible with the current immersive animation context, identify drag-and-drop selection behaviors and operation sequences, and generate target data packages by combining them with a pre-built decision coordinate system. Transmission Feedback Module: Based on the target data packet, it performs multi-end data alignment, calculates the variance under the same decision node, and generates a personal ACP preference report containing consensus items and highly concerned conflict items.

[0006] Furthermore, the multidimensional data includes user-inputted identity identifiers, clinical status, and physiological indicators; the identity identifiers include at least the user ID, the clinical status includes any one of healthy, sub-healthy, stable period of chronic disease, or end-stage, and the physiological indicators include at least the resting heart rate; Furthermore, rule matching is introduced to generate immersive animated scenarios adapted to user profiles, including: Call the preset progressive script library, including the first script for configuring the first level, the second script for configuring the second level, and the third script for configuring the third level. Simultaneously, all script fragments in the progressive script library are analyzed, and information entropy and stylization features are extracted to construct a two-dimensional space. All script fragments are clustered and stored in the two-dimensional space to form a content attribute space. User feature vectors are parsed from user profiles. Through orthogonal decomposition, the first, second, and third components are extracted. Three rule matching operations are performed in the content attribute space to lock the corresponding scripts and their coordinate points. Combined with the projection direction of the third component onto the horizontal axis in two-dimensional space, a preset graphics rendering engine is called to output an immersive animation scenario.

[0007] Furthermore, three rule matching operations are performed in the content attribute space, including: First rule matching: Compare the deviation of the first component from the preset benchmark value, set the gradient range of the deviation, if the deviation exceeds the gradient range of ±wc%, map to the low entropy value range to lock the first script; if the deviation does not exceed the gradient range of ±wc%, then perform second rule matching. Secondary rule matching: Evaluate the weight of the second component, filter out cases indicating high weight, map them to the high entropy value range to lock the second script; if the weight value indicates a low weight state, then perform tertiary rule matching; Three-stage rule matching: directly mapping to the intermediate information entropy interval to lock the second script.

[0008] Furthermore, by implementing scenario-based guidance, a comprehensive guidance index is obtained, including: Based on the touch trajectory, the touch withdrawal frequency and operation response delay are extracted and normalized into the first guiding factor; Based on physiological indicators, the deviation of the current physiological indicators from the preset benchmark value is extracted and normalized into a second guiding factor; the first guiding factor and the second guiding factor are weighted and summed to obtain the comprehensive guiding index. At the same time, based on the first guiding factor, a hierarchical jump strategy is executed. If it is determined that the current script corresponds to the third level, a visual remapping is performed.

[0009] Furthermore, a hierarchical jump strategy is executed based on the first guiding factor, including: If the first guidance factor is less than the preset smoothness threshold, it is determined that the current script level is passed, the state machine is triggered to jump to the next level, and the playback pointer of the scenario guidance sequence is automatically jumped to the script entry point corresponding to the next level. If the first guiding factor is greater than the preset smoothness threshold, it is determined that the user has an interaction obstacle and has not passed the current level. A rollback instruction is triggered, the low information entropy script of the current level is called for replay, and the jump to the next level is suppressed.

[0010] Furthermore, the implementation of selection decision-making strategies includes: The system identifies the current level of the immersive animation context, calls the hierarchical card database, determines the wish cards with the same attribute tags as that level, and loads them into the selection area; each wish card has a unique identifier. The decision coordinate system is configured based on attribute labels and divided into quadrant regions with different attribute labels; Users can drag and drop wish cards into the decision coordinate system, and obtain the drag landing coordinates and the timestamp of selection for each selected wish card in real time. The position weight is determined based on the drag-and-drop coordinates, and the time weight is determined based on the timestamp. The position weight and time weight are weighted and fused to generate the final weight of the wish card. The target data packet is encapsulated and generated based on the unique identifier of the selected wish card, its quadrant region, and the final weight.

[0011] Furthermore, a personal ACP preference report is generated, including consensus items and highly concerned conflict items, including: Each unique identifier is marked as a decision node; at the same time, the user's identity identifier is called to extract the user ID, and a corresponding collaboration matrix is ​​established with the user ID as the row vector and the decision node as the column vector; Traverse the collaboration matrix and fill the corresponding cells of the collaboration matrix with the final weights based on the correspondence between user IDs and decision nodes. For a decision node that a user has not selected, the system automatically fills the cell corresponding to that user with zero values ​​to complete the sparse alignment of multi-terminal data. For each decision node, extract the set of all non-zero final weights under that column vector, calculate the variance, and compare the variance with a preset variance threshold: if the variance is greater than the preset variance threshold, the decision node is determined to be a high-concern conflict item; if the variance is less than or equal to the preset variance threshold, the decision node is determined to be a consensus item.

[0012] Secondly, this application provides a data analysis method based on ACP-guided scenario interaction, the method including: collecting multidimensional data of users, configuring user profiles, introducing rule matching, and generating immersive animation scenarios adapted to user profiles; The interactive experience utilizes immersive animated scenarios, capturing touch trajectories and physiological indicators during the interaction to guide the scenario and obtain a comprehensive guidance index. Based on this comprehensive guidance index, fluctuation trends are identified, and the scenario progression rate is dynamically adjusted. Based on the comprehensive guidance index, the selection decision strategy is executed, the wish cards that are compatible with the current immersive animation context are loaded, and the drag-and-drop selection behavior and operation sequence are identified. Combined with the pre-built decision coordinate system, the target data package is generated. Based on the target data packet, perform multi-terminal data alignment, calculate the variance under the same decision node, and generate a personal ACP preference report containing consensus items and highly concerned conflict items.

[0013] (III) Beneficial Effects This invention provides a data analysis system and method based on ACP-guided scenario interaction, which has the following beneficial effects: This invention quantifies multidimensional user data, configures user profiles, and introduces rule matching to map user profiles to content attribute space. Through three rule matching steps, the best script is selected from a progressive script library. This enables the automatic establishment of script access based on user profiles without manual intervention, achieving rapid generation of immersive animation scenarios, ensuring decision-making efficiency and improving timeliness. This invention quantifies the deviation characteristics of touch trajectory and physiological indicators into a first guiding factor and a second guiding factor, and generates a comprehensive guiding index, which reduces the complexity of analyzing massive amounts of unstructured interactive data. At the same time, based on the comprehensive guiding index, a fluctuation trend recognition logic is constructed to dynamically adjust the scenario advancement rate, which reduces the misjudgment rate caused by user mis-touch or physiological noise, and improves the robustness and adaptive control accuracy of the system. This invention implements a selection decision-making strategy based on a comprehensive guidance index. In this process, by integrating the drag-and-drop coordinates of the wish card with the selected timestamp, the position weight and time weight are determined, and the final weight is generated. This provides scientific data support for subsequent family consensus comparison, thereby reducing communication costs in medical implementation and improving the matching degree of care plans. This invention quantifies the discrete intention data among family members into statistical variance by executing multi-end data alignment and variance calculation logic, and generates consensus terms and high-concern conflict terms based on this, thereby improving the system's accuracy in identifying potential family intentions, enabling automatic intervention in high-concern conflict terms, and improving the accuracy of ACP guidance and expression. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the modules of the present invention; Figure 2 This is a schematic diagram of the steps of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] Example 1: This embodiment of the invention provides a data analysis system based on ACP-guided scenario interaction; Figure 1 This is a schematic diagram of the modules of the present invention. The project aims to develop a digital ACP communication system called Heart-to-Heart Communication Card. Through innovative design that incorporates localization, gamification, and digitalization, it constructs an ACP communication platform that integrates scenario guidance, personal reflection, family collaboration, and professional support. The core of the project is to address the pain point of communication difficulties and provide individuals, families, and nursing professionals with a friendly, effective, and scalable ACP communication solution, thereby promoting the improvement of end-of-life care quality.

[0017] Please see Figure 1The system includes: a profile building module, a scenario guidance module, a selection and decision-making module, and a transmission and feedback module, and the profile building module, scenario guidance module, selection and decision-making module, and transmission and feedback module are interconnected; it provides ACP communication training workshops and tool kits for hospitals, elderly care institutions, and community health service centers, and embeds this product into palliative care clinical pathways and health promotion activities; The following is an explanation of each module involved: User profile building module: Collects multidimensional user data, configures user profiles, introduces rule matching, and generates immersive animation scenarios adapted to user profiles; The multidimensional data includes physiological indicators, clinical status, and identity identifiers input by the user. Physiological indicators: including the user's resting heart rate and skin conductance response data; these are read and stored as the user's personalized physiological indicators by connecting to the user's wearable biomonitoring device via a short-range wireless communication interface, such as Bluetooth. Clinical status: This includes the user's clinical stage, which is healthy, sub-healthy, stable stage of chronic disease, or terminal stage. The user terminal establishes an encrypted data channel with the third-party electronic medical record system (EMR) to request the user's main diagnostic records and disease stage data. If the data channel request times out or no data is returned, it automatically switches to an adaptive questionnaire mode, presenting a standardized physical function status scoring table on the interface. Based on the obtained diagnostic records or scoring results, the user's clinical status is classified into one of four categories: healthy, sub-healthy, stable stage of chronic disease, or terminal stage. Identity identifiers include user ID, gender, age, cultural background tags, and education level. The user terminal renders a structured information input form through a human-computer interaction interface to obtain the user's input education level and cultural background tags. At the same time, the user terminal's OCR module is called to scan the user's identity document, automatically extract and verify the user's age and gender. The above data is then mapped to obtain the identity identifier dataset. Features are extracted from multidimensional data and fused to generate unique user feature vectors to build user profiles for subsequent model input and algorithm calculations. Introducing rule-based matching to generate immersive animated scenarios tailored to user profiles, including: Call the preset progressive script library, including the first script for configuring the first level, the second script for configuring the second level, and the third script for configuring the third level. Simultaneously, a two-dimensional space is configured. By parsing all script fragments in the progressive script library, the information entropy of each script fragment is extracted and used as the vertical axis of the two-dimensional space. The first script is mapped to the low entropy range, the second script to the medium entropy range, and the third script to the high entropy range. An image recognition algorithm is used to extract the stylistic features of each script fragment and use them as the horizontal axis of the two-dimensional space to generate unique coordinates for each script fragment in the two-dimensional coordinate space. Based on the unique coordinates, all script fragments are clustered and stored in the two-dimensional space to form a content attribute space. It should be noted that the two-dimensional space includes information entropy and stylization features. Information entropy is used to measure the density of medical entity words and the degree of explicitness of death-related semantics in the multimedia script. Stylization features are used to measure the cultural symbol attributes of visual elements in the multimedia script, and are continuously quantified from modern minimalist style to traditional humanistic style. Based on user profiles, user feature vectors are parsed out. Through orthogonal decomposition, the first, second, and third components are extracted, and three rule matching operations are performed in the content attribute space to locate the corresponding scripts. The first component is used to represent the user's current real-time physiological indicators; the second component is used to represent the user's clinical state; and the third component is used to represent the user's cultural cognition. First rule matching: Compare the deviation of the first component from the preset benchmark value, set the gradient range of the deviation, if the deviation exceeds the gradient range of ±wc%, map to the low entropy value range to lock the first script; if the deviation does not exceed the gradient range of ±wc%, then perform second rule matching. Preset baseline values: Read historical mean resting heart rate and historical skin conductance response data; perform sliding window filtering and outlier removal on the above data, retain valid physiological indicators, calculate their arithmetic mean, and store it as the user's personalized physiological baseline value indicators; Deviation: Compare the deviation of the first component with the preset benchmark value. That is, calculate the difference between any physiological indicator and the preset benchmark value, and perform normalization to obtain the corresponding deviation. If the deviation of a physiological indicator exceeds the gradient range of ±wc%, it is mapped to the low entropy value range. Secondary rule matching: Evaluate the weight of the second component, filter out cases indicating high weight, map them to the high entropy value range to lock the second script; if the weight value indicates a low weight state, then perform tertiary rule matching; The evaluation of the weight of the second component includes: reading the second component, which is a feature vector representing the user's pre-stage state. The system has a built-in weight mapping table for state stages. By mapping the second component into numerical weight values, the weight value corresponding to the terminal stage is set to the maximum, and the weight value corresponding to health is set to the minimum, satisfying: weight value of health < weight value of sub-health < weight value of stable period of chronic disease < weight value of terminal stage; comparing the weight value with a preset weight threshold: if the weight value is greater than or equal to the weight threshold, it is determined that the current state is high-weight, and it is mapped to a region with high information entropy dimension in the content attribute space, and the third script in that region is locked; if the weight value is less than the weight threshold, it is determined that the current state is low-weight, and the process proceeds to three rule matching steps. Three-stage rule matching: directly mapping to the intermediate information entropy interval to lock the second script; Based on the coordinates of the first, second, and third scripts, and combined with the projection direction of the third component on the horizontal axis, a preset graphics rendering engine is invoked to output an immersive animation scenario. This includes: retrieving the locked first, second, or third script from the content attribute space and using it as the scene content base; calculating the projection vector of the third component on the horizontal axis; if the projection vector points in the positive direction of the horizontal axis, a preset traditional humanistic graphics rendering engine is invoked to load ink painting or natural landscape textures for rendering and compositing the script; conversely, if the projection vector points in the negative direction of the horizontal axis, a preset modern minimalist graphics rendering engine is invoked to load geometric or data chart textures for rendering and compositing the script; the final output is an immersive animation scenario. The significance of the above analysis lies in the fact that by quantifying multidimensional user data, configuring user profiles, and introducing rule matching, script access can be automatically established based on user profiles without human intervention, enabling the rapid generation of immersive animation scenarios, thus ensuring decision-making efficiency and improving timeliness.

[0018] Contextual Guidance Module: Uses immersive animated contexts for interaction, and acquires touch trajectories and physiological indicators during the interaction to perform contextual guidance and obtain a comprehensive guidance index; Based on the comprehensive guidance index, it identifies fluctuation trends and dynamically adjusts the contextual progression rate; Touch trajectory: Records the position of the user's finger touching the screen in the APP. For example, the coordinates are recorded 60 times per second, which is represented as a set of spatiotemporal coordinate points: G={(x1, y1, t1), (x2, y2, t2)……}; where (x1, y1) and (x2, y2) represent the pixel coordinates on the screen, and t1 and t2 represent the first and second timestamps; Execute scenario-based guidance to obtain a comprehensive guidance index, including: During the current script playback, create a sliding time window: Based on touch trajectories, the cumulative number of times touch points are deselected in the screen coordinate system is extracted to determine the touch withdrawal frequency, which is used to characterize the user's operation jitter. Simultaneously, the hovering duration of touch points in a preset interaction area is extracted to determine the operation response delay, which is used to characterize the user's operation response delay. A linear normalization algorithm is used to normalize the cumulative number of times and the hovering duration to generate a first guiding factor. This factor is calculated by calculating the arithmetic mean of the discrete distribution of touch points and the hovering duration in a specific interaction area, which characterizes the user's behavioral resistance level during the interaction process. Simultaneously, a hierarchical jump strategy is executed based on the first guiding factor, including: If the first guiding factor is less than the preset smoothness threshold, the current script level is determined to be passed, the state machine is triggered to jump to the next level, and the playback pointer of the scenario guiding sequence is automatically jumped to the script entry point corresponding to the next level. If the first guiding factor is greater than the preset smoothness threshold, it is determined that the user has an interaction obstacle and the current level has not been passed. A rollback instruction is triggered, the low information entropy script of the current level is called for replay, and the jump to the next level is suppressed. Preset smoothness threshold: Based on historical statistics, historical first guiding factor data is collected, and datasets with first guiding factors less than the preset smoothness threshold are extracted. The mean and standard deviation of the dataset are obtained, and the mean and a certain multiple of the standard deviation are used as the smoothness threshold. It should be noted that the certain multiple is between 2 and 3, and the smoothness threshold is only a reference indicator. The specific value should be determined according to the actual situation. If the current script corresponds to the third level, the system scans the original medical scene material of the third level frame by frame and uses object detection algorithms to identify semantic objects. Among them, specific semantic objects include medical devices, ward facilities and emergency equipment. The system calls the pre-stored graph texture library, which stores element textures that are adapted to the user profile. The system performs visual remapping processing to replace the identified semantic objects with the corresponding element textures in real time, generating a de-medicalized rendering frame. Simultaneously, the system calls the built-in background sound track, which is a preset white noise or soothing melody; it calls the current comprehensive guidance index and synthesizes binaural beat signals for superposition; by establishing a negative correlation mapping function between the frequency difference of binaural beats and the comprehensive guidance index: if the comprehensive guidance index increases, the frequency difference of binaural beats is automatically adjusted to converge towards the Alpha band, such as 8 to 13 Hz or the Theta band, such as 4 to 8 Hz, and the user's brainwave frequency is induced to decrease synchronously through the audio output interface; Based on physiological indicators, and reading the deviation of the user's pre-stored preset benchmark values, including the deviation of the user's resting heart rate from the corresponding preset benchmark value and the deviation of the skin conductance response data from the corresponding preset benchmark value, the arithmetic mean of the two deviations is calculated using a linear normalization algorithm to obtain the second guidance factor; corresponding weight coefficients are assigned to the first guidance factor and the second guidance factor, and the comprehensive guidance index is obtained by weighted summation; The weighting coefficients are sourced as follows: The weighting coefficient for the first guiding factor is based on the user profile's identity data and clinical status data. If the data identifies a user as elderly or prone to accidental touches, and determines that their touch trajectory contains high physical noise, a lower weighting coefficient is assigned; conversely, the system outputs a higher first weighting coefficient. The weighting coefficient for the second guiding factor is based on the sampling continuity and signal-to-noise ratio of real-time monitored physiological indicators. The signal-to-noise ratio (SNR) is mapped and calculated using the Sigmoid activation function: when the SNR is higher than the standard SNR threshold, the physiological indicators are deemed to have reference value, and the system outputs a higher weight coefficient; if the SNR is lower than the preset standard SNR threshold, the physiological data is deemed to be distorted, and the system outputs a weight coefficient close to zero. Based on a comprehensive guiding index, fluctuation trends are identified, and the scenario advancement rate is dynamically adjusted, including: Based on the comprehensive guiding index, a corresponding time series curve is constructed to determine the slope of change at the current moment, thereby identifying the fluctuation trend. If the slope of change is greater than 0 and the absolute value exceeds the preset slope threshold, it is determined to be a positive fluctuation trend, indicating that the user's behavioral resistance or physiological load is increasing significantly; at the same time, a damping adjustment command is triggered to reduce the scenario advancement rate in the opposite direction according to the magnitude of the slope of change. If the slope of change is equal to 0, it is determined to be a steady trend, and the current rate of progress of the scenario is maintained; If the slope of change is less than 0, it is determined to be a downward fluctuation trend. At the same time, a recovery command is triggered to gradually increase the scenario advancement rate. The significance of the above analysis lies in the following: During the interaction process, the deviation characteristics of touch trajectory and physiological indicators are analyzed and quantified into the first and second guiding factors, and a comprehensive guiding index is generated, which reduces the complexity of analyzing massive amounts of unstructured interaction data; at the same time, based on the comprehensive guiding index, a fluctuation trend recognition logic is constructed to dynamically adjust the scenario advancement rate, which reduces the misjudgment rate caused by user's single behavior accidental touch or physiological noise, and improves the robustness and adaptive control accuracy of the system.

[0019] Selection Decision Module: Based on the comprehensive guidance index, the module executes selection decision strategies, loads wish cards that are compatible with the current immersive animation context, identifies drag-and-drop selection behaviors and operation sequences, and generates target data packages by combining them with a pre-built decision coordinate system. Implementing selection decision-making strategies includes: The system identifies the current level of the immersive animation context, calls the hierarchical card database, retrieves wish cards with the same attribute tags as that level, and loads the wish cards into the selection area to achieve semantic adaptation between the wish cards and the guiding context; each wish card has a unique identifier. In addition, for user-defined text cards, NLP is used to extract text keywords, calculate the cosine similarity between them and the vectors of each dimension in the preset dimension word vector library, and determine the attribute label of the custom text card based on the maximum similarity. At the same time, the decision coordinate system is configured based on the attribute labels, and the coordinate system is divided into quadrant regions for different attribute labels. For example, the attribute label usually takes the value 4, which corresponds to 4 quadrant regions. That is, when the coordinates of the wish card fall within the coordinate range of a certain quadrant, the system automatically assigns the attribute corresponding to that quadrant to the wish card. Users can drag and drop wish cards into the decision coordinate system, and obtain the drag landing coordinates and the timestamp of selection for each selected wish card in real time. Position weight is determined based on the drag-and-drop landing point coordinates: the Euclidean distance of the wish card relative to the geometric center of its quadrant is calculated, and the position weight is generated using the inverse distance proportionality function; it should be noted that the closer the landing point of the wish card is to the quadrant center, the higher the generated position weight value. Determining time weights based on timestamps: Calculate the time difference between the selected timestamp and the start time of the selection decision module, introduce a time decay function, and generate time weights; it should be noted that the time decay function is a monotonically decreasing function, which means that the smaller the time difference, the higher the time weight; The final weight of the Wish Card is generated by weighting and combining the position weight and the time weight; Based on the unique identifier of the selected wish card, its quadrant region, and its final weight, the data is encapsulated according to a preset data structure to generate a target data packet for subsequent data storage and transmission. The significance of the above analysis lies in the fact that by implementing a selection decision-making strategy based on a comprehensive guidance index, and by integrating the drag-and-drop coordinates of the wish cards with the selected timestamp, the position weight and time weight are determined, and the final weight is generated. This provides scientific data support for subsequent family consensus comparison, thereby reducing communication costs in medical implementation and improving the matching degree of care plans.

[0020] Transmission feedback module: Based on the target data packet, it performs multi-end data alignment, calculates the variance under the same decision node, and generates a personal ACP preference report containing consensus items and highly concerned conflict items; This includes generating a personal ACP preference report containing consensus items and highly concerned conflict items, including: Parse the target data packet and extract the unique identifier and final weight of all wish cards contained therein; Each unique identifier is marked as an independent decision node; at the same time, the user's identity identifier is called to extract the user ID, and a corresponding collaboration matrix is ​​established with the user ID as the row vector and the decision node as the column vector. Traverse the collaboration matrix and fill the corresponding cells of the collaboration matrix with the final weights based on the correspondence between user IDs and decision nodes. For a decision node that a user has not selected, the system automatically fills the cell corresponding to that user with zero values ​​to complete the sparse alignment of multi-terminal data. For each decision node, extract the set of all non-zero final weights under that column vector, calculate the variance, and compare the variance with a preset variance threshold. If the variance is greater than the preset variance threshold, the decision node is determined to be a high-concern conflict item, indicating that family members have not reached a consensus on the corresponding wish card and there is a significant disagreement. At the same time, the high-concern conflict item is packaged into an independent discussion task package, distributed to each associated user terminal, and the audio acquisition component is automatically activated to establish a multi-terminal real-time voice communication channel until the variance of the high-concern conflict item after resubmission is less than or equal to the preset variance threshold. If the variance is less than or equal to the preset variance threshold, the decision node is determined to be a consensus term, indicating that family members have similar wishes on the corresponding wish cards; The preset variance threshold is used to identify the optimal conflict determination boundary through cross-validation, and the continuous variance values ​​are accurately divided into consensus terms that are high-concern conflict terms. The significance of the above analysis lies in the following: by executing multi-end data alignment and variance calculation logic, the discrete intention data among family members are quantified into statistical variance, and consensus terms and high-concern conflict terms are generated based on this, which improves the system's accuracy in identifying potential family intentions, so as to realize automatic intervention for high-concern conflict terms, provide precise intervention navigation basis for clinical medical staff, and further improve the accuracy of ACP guidance and expression.

[0021] In the clinical nursing field, as a standardized communication aid, it is applied in palliative care wards, geriatric departments, and oncology departments of tertiary hospitals, helping nurses initiate ACP (Aspect-Care Practice) dialogues more efficiently and humanely, thereby improving nursing quality and patient satisfaction. Through nursing education and training: In the community and public health sectors, it is promoted to healthy and sub-healthy middle-aged and elderly groups through community health service centers and elderly care institutions, serving as part of healthy aging education, proactively prompting public reflection on life planning and changing social perceptions. Through data value and research output: Under the premise of strictly protecting user privacy, it accumulates first-hand data on the ACP preferences of the Chinese population, providing valuable resources for government decision-making and academic research, and continuously promoting the development of a localized palliative care system. Business model expansion: In the early stages, it can combine public welfare and commercial aspects through government procurement and hospital / institutional procurement models. In the long term, it can explore value-added services (such as personalized reports and in-depth consultations) for the general public to achieve sustainable development of the project.

[0022] Example 2: This embodiment of the invention provides a data analysis method based on ACP-guided scenario interaction; Figure 2 This is a schematic diagram of the steps of the present invention; please refer to [link / reference]. Figure 2 The method includes the following steps: Collect multidimensional user data, configure user profiles, introduce rule matching, and generate immersive animated scenarios that are adapted to user profiles; The interactive experience utilizes immersive animated scenarios, capturing touch trajectories and physiological indicators during the interaction to guide the scenario and obtain a comprehensive guidance index. Based on this comprehensive guidance index, fluctuation trends are identified, and the scenario progression rate is dynamically adjusted. Based on the comprehensive guidance index, the selection decision strategy is executed, the wish cards that are compatible with the current immersive animation context are loaded, and the drag-and-drop selection behavior and operation sequence are identified. Combined with the pre-built decision coordinate system, the target data package is generated. Based on the target data packet, perform multi-terminal data alignment, calculate the variance under the same decision node, and generate a personal ACP preference report containing consensus items and highly concerned conflict items.

[0023] In the application, the various calculation methods mentioned all involve dimensionless calculations, and the calculation methods are derived from software simulations based on a large amount of data to obtain the most recent real-world situation. These methods are set by those skilled in the art according to the actual situation.

[0024] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0025] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0026] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A data analysis system based on ACP-guided scenario interaction, characterized in that, The system includes: User profile building module: used to collect multidimensional user data, configure user profiles, introduce rule matching, and generate immersive animation scenarios adapted to user profiles; Contextual Guidance Module: Uses immersive animated contexts for interaction, and acquires touch trajectories and physiological indicators during the interaction to perform contextual guidance and obtain a comprehensive guidance index; Based on the comprehensive guidance index, it identifies fluctuation trends and dynamically adjusts the contextual progression rate; Selection Decision Module: This module is used to execute selection decision strategies based on a comprehensive guidance index, load wish cards that are compatible with the current immersive animation context, identify drag-and-drop selection behaviors and operation sequences, and generate target data packages by combining them with a pre-built decision coordinate system. Transmission Feedback Module: Based on the target data packet, it performs multi-end data alignment, calculates the variance under the same decision node, and generates a personal ACP preference report containing consensus items and highly concerned conflict items.

2. The data analysis system based on ACP-guided scenario interaction according to claim 1, characterized in that, The multidimensional data includes user-inputted identity identifiers, clinical status, and physiological indicators; the identity identifiers include at least a user ID, the clinical status includes any one of healthy, sub-healthy, stable period of chronic disease, or end-stage, and the physiological indicators include at least resting heart rate.

3. A data analysis system based on ACP-guided scenario interaction according to claim 2, characterized in that, The introduction of rule matching to generate immersive animation scenarios adapted to user profiles includes: Call the preset progressive script library, including the first script for configuring the first level, the second script for configuring the second level, and the third script for configuring the third level. Simultaneously, all script fragments in the progressive script library are analyzed, and information entropy and stylization features are extracted to construct a two-dimensional space. All script fragments are clustered and stored in the two-dimensional space to form a content attribute space. User feature vectors are parsed from user profiles. Through orthogonal decomposition, the first, second, and third components are extracted. Three rule matching operations are performed in the content attribute space to lock the corresponding scripts and their coordinate points. Combined with the projection direction of the third component onto the horizontal axis in two-dimensional space, a preset graphics rendering engine is called to output an immersive animation scenario.

4. A data analysis system based on ACP-guided scenario interaction according to claim 3, characterized in that, The three-stage rule matching in the content attribute space includes: First rule matching: Compare the deviation of the first component from the preset benchmark value, set the gradient range of the deviation, if the deviation exceeds the gradient range of ±wc%, map to the low entropy value range to lock the first script; if the deviation does not exceed the gradient range of ±wc%, then perform second rule matching. Secondary rule matching: Evaluate the weight of the second component, filter out cases indicating high weight, map them to the high entropy value range to lock the second script; if the weight value indicates a low weight state, then perform tertiary rule matching; Three-stage rule matching: directly mapping to the intermediate information entropy interval to lock the second script.

5. A data analysis system based on ACP-guided scenario interaction according to claim 4, characterized in that, The execution scenario guidance yields a comprehensive guidance index, including: Based on the touch trajectory, the touch withdrawal frequency and operation response delay are extracted and normalized into the first guiding factor; Based on physiological indicators, the deviation of the current physiological indicators from the preset benchmark value is extracted and normalized into a second guiding factor; the first guiding factor and the second guiding factor are weighted and summed to obtain the comprehensive guiding index. At the same time, based on the first guiding factor, a hierarchical jump strategy is executed. If it is determined that the current script corresponds to the third level, a visual remapping is performed.

6. A data analysis system based on ACP-guided scenario interaction according to claim 5, characterized in that, The hierarchical jump strategy based on the first guiding factor includes: If the first guidance factor is less than the preset smoothness threshold, it is determined that the current script level is passed, the state machine is triggered to jump to the next level, and the playback pointer of the scenario guidance sequence is automatically jumped to the script entry point corresponding to the next level. If the first guiding factor is greater than the preset smoothness threshold, it is determined that the user has an interaction obstacle and has not passed the current level. A rollback instruction is triggered, the low information entropy script of the current level is called for replay, and the jump to the next level is suppressed.

7. A data analysis system based on ACP-guided scenario interaction according to claim 6, characterized in that, The execution selection decision strategy includes: The system identifies the current level of the immersive animation context, calls the hierarchical card database, determines the wish cards with the same attribute tags as that level, and loads them into the selection area; each wish card has a unique identifier. The decision coordinate system is configured based on attribute labels and divided into quadrant regions with different attribute labels; Users can drag and drop wish cards into the decision coordinate system, and obtain the drag landing coordinates and the timestamp of selection for each selected wish card in real time. The position weight is determined based on the drag-and-drop coordinates, and the time weight is determined based on the timestamp. The position weight and time weight are weighted and fused to generate the final weight of the wish card. The target data packet is encapsulated and generated based on the unique identifier of the selected wish card, its quadrant region, and the final weight.

8. A data analysis system based on ACP-guided scenario interaction according to claim 7, characterized in that, The generation of a personal ACP preference report, which includes consensus items and highly concerned conflict items, includes: Parse the target data packet and extract the unique identifier and final weight of all wish cards contained therein; Each unique identifier is marked as a decision node; at the same time, the user's identity identifier is called to extract the user ID, and a corresponding collaboration matrix is ​​established with the user ID as the row vector and the decision node as the column vector; Traverse the collaboration matrix and fill the corresponding cells of the collaboration matrix with the final weights based on the correspondence between user IDs and decision nodes. For a decision node that a user has not selected, the system automatically fills the cell corresponding to that user with zero values ​​to complete the sparse alignment of multi-terminal data. For each decision node, extract the set of all non-zero final weights under that column vector, calculate the variance, and compare the variance with a preset variance threshold: if the variance is greater than the preset variance threshold, the decision node is determined to be a high-concern conflict item; if the variance is less than or equal to the preset variance threshold, the decision node is determined to be a consensus item.

9. A data analysis method based on ACP-guided scenario interaction, characterized in that, The method includes: collecting multidimensional data of users, configuring user profiles, introducing rule matching, and generating immersive animation scenarios adapted to user profiles; The interactive experience utilizes immersive animated scenarios, capturing touch trajectories and physiological indicators during the interaction to guide the scenario and obtain a comprehensive guidance index. Based on this comprehensive guidance index, fluctuation trends are identified, and the scenario progression rate is dynamically adjusted. Based on the comprehensive guidance index, the selection decision strategy is executed, the wish cards that are compatible with the current immersive animation context are loaded, and the drag-and-drop selection behavior and operation sequence are identified. Combined with the pre-built decision coordinate system, the target data package is generated. Based on the target data packet, perform multi-terminal data alignment, calculate the variance under the same decision node, and generate a personal ACP preference report containing consensus items and highly concerned conflict items.

Images