A Visual Assessment Method and System for Rehabilitation Services

Abstract

This invention relates to the field of data visualization and integration technology, specifically to a method and system for visually evaluating rehabilitation services. The method includes the following steps: acquiring and standardizing multi-dimensional indicator data; establishing a normalized matrix mapping parameter; and generating a set of visual evaluation charts for rehabilitation services. In this invention, by constructing a unified numbering set covering multi-dimensional indicators, effective organization and standardized management of data sources are achieved. Data fusion compatibility and accuracy are improved through field similarity comparison and unit conversion. Normalization calculations and grade grouping operations for values ​​such as service usage, resource allocation, job distribution, and fiscal expenditure ensure that various indicators are compared under a unified standard. Visual mapping of core parameters such as resource ratio, efficiency ratio, and coverage ratio is used. Interactive components and layer switching enhance the dynamic response and multi-view switching capabilities of the graphical interface, solving the problems of insufficient structure, single result presentation format, and lack of real-time performance in traditional evaluations.

Description

Technical Field

[0001] This invention relates to the field of data visualization and integration technology, and in particular to a method and system for visual assessment of rehabilitation services. Background Technology

[0002] Image analysis technology encompasses techniques for extracting, recognizing, processing, and analyzing information from static images or dynamic videos. These include image preprocessing, feature extraction, object detection, image segmentation, pattern recognition, and structured image representation, and are widely applied in various scenarios such as medical diagnosis, industrial inspection, and behavior recognition. Traditional methods for visualizing rehabilitation service assessments refer to methods used to acquire, analyze, and present rehabilitation assessment data during the rehabilitation service process. These methods typically rely on manual reporting and paper records for data collection. Assessment indicators are relatively scattered and primarily qualitative, making it difficult to form structured data. The assessment process mainly uses manual comparison and single-dimensional statistics for result analysis. Assessment conclusions are primarily presented in tabular or textual form, lacking visual representation of key indicators such as service efficiency, coverage, and resource allocation, and lacking multi-dimensional data fusion and real-time processing mechanisms.

[0003] Current technologies for rehabilitation service assessment rely on manual reporting and paper records, resulting in time-consuming and inaccurate data acquisition. Assessment indicators are often fragmented and primarily qualitative, making it difficult to structure the data. Analysis primarily relies on manual comparison and static statistics of a single dimension, lacking the ability to comprehensively analyze the correlations between multidimensional data. Assessment results are often presented in tabular or textual form, failing to intuitively express the relationships and trends between key elements. Furthermore, the lack of real-time updates and dynamic data response mechanisms leads to delayed assessment results and limited effectiveness, failing to provide accurate assessments of service efficiency, resource coverage, and allocation, and hindering effective support for optimizing rehabilitation services. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a visual assessment method for rehabilitation services.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for visually assessing rehabilitation services, comprising the following steps:

[0006] S1: Obtain the dimension indicator entries in the TRIC and FRAME structures and collect the data sources. Identify the indicator codes, measurement units, and level classifications in the data sources. Combine the indicator dimensions to set the collection methods and evaluation thresholds respectively, and generate a set of rehabilitation dimension indicator numbers.

[0007] S2: Based on the rehabilitation dimension indicator number set, obtain the structured questionnaire platform data and reporting database of rehabilitation service institutions in each region, detect the coding method, time format and unit type in each data, perform label comparison and unit conversion operations through field similarity, and generate a unified field sequence table;

[0008] S3: Based on the service usage count, resource allocation quantity, job function distribution, fiscal expenditure details and institution level identifier in the unified field sequence table, obtain the numerical distribution range under the same time dimension, perform weighted proportional allocation on the resource allocation quantity and job function distribution, perform standardized interval normalization on the service usage count and fiscal expenditure details, and synchronously group all normalization results according to the rehabilitation service institution level identifier to generate a grouped weighted normalization matrix.

[0009] S4: Based on the grouped weighted normalization matrix, calculate the resource ratio, utilization efficiency ratio and job coverage ratio of each group, extract the city code and connect it to the administrative region coordinate dictionary, set the starting point of plane rendering according to the distribution order of rehabilitation service institutions, perform parameter mapping, and generate a graphic coordinate mapping parameter table.

[0010] S5: Based on the graphic coordinate mapping parameter table, set the time dimension scrolling component and the comparison item selection control, load all graphics into the main view area of ​​the interactive interface, and generate a group of visual assessment charts for rehabilitation services.

[0011] As a further aspect of the present invention, the dimension indicators specifically refer to leadership and governance, fundraising mechanisms, human resource allocation, infrastructure conditions, and service accessibility and quality.

[0012] The data sources include, but are not limited to, rehabilitation institution system management documents, financial expenditure records, human resource allocation files, facility equipment ledgers, and outpatient usage ledgers;

[0013] Specifically, the execution parameter mapping refers to mapping the resource ratio to the radar coordinate system, mapping the utilization efficiency ratio to the bar chart coordinate axis, and mapping the job coverage ratio to the heat map color scale.

[0014] The rehabilitation dimension indicator number set includes an indicator number structure, a measurement unit system, level classification rules, data collection method parameters, and evaluation threshold settings. The unified field sequence table includes a unified coding label set, a unified time format set, a unified unit conversion table, field semantic classification labels, and field matching confidence scores. The grouped weighted normalization matrix includes a time dimension grouping table, a resource allocation weight matrix, a job function ratio matrix, a service standardization usage value set, a fiscal expenditure normalization dataset, and an institution level classification label. The graphic coordinate mapping parameter table includes a radar chart ratio parameter set, a bar chart efficiency parameter set, a heat map coverage parameter set, an administrative coordinate reference table, and a rendering starting point sequence. The rehabilitation service visualization assessment chart set includes a radar chart layer, a bar chart layer, a heat map layer, a layer switching control, a time scrolling component, a comparison selection control, and a main view area display box.

[0015] As a further aspect of the present invention, the steps for obtaining the rehabilitation dimension indicator number set are as follows:

[0016] S111: Obtain five types of data sources: rehabilitation institution system management documents, financial expenditure records, human resource allocation files, facility equipment ledgers, and outpatient usage ledgers. Perform structural parsing on the record fields in each data source, extract the indicator names, coding formats, measurement units, and level classification parameters from the fields, and map them one by one to the dimensions of leadership and governance, financing mechanisms, human resource allocation, infrastructure conditions, and service accessibility and quality according to the field names. Perform deduplication on the coding items with duplicate fields in the mapping relationship to generate a dimension indicator field mapping set.

[0017] S112: Based on the dimension indicator field mapping set, extract the measurement unit and level classification parameters of each indicator item, establish an indicator collection structure template, and divide the data source categories of the fields into three collection methods: document collection, structured record extraction, and ledger statistics. Match the collection methods with the corresponding field structure templates one by one, and attach the collection methods to the indicator coding items to generate an indicator collection structure combination table.

[0018] S113: Based on the collection method and field level classification parameters of each indicator in the indicator collection structure combination table, read the measurement unit and level classification value range pointed to by each indicator, compare and calculate the level parameter value with the reference level threshold, mark the judgment threshold parameter corresponding to each level interval, aggregate and configure the indicator code, collection method and judgment threshold, and generate a set of rehabilitation dimension indicator numbers.

[0019] As a further aspect of the present invention, the step of obtaining the unified field sequence table specifically includes:

[0020] S211: Obtain structured questionnaire platform data and reporting database of rehabilitation service institutions in various regions, extract field content, field code and field unit in each dataset, cross-match field nouns with indicator field names in the rehabilitation dimension indicator number set, identify whether there are naming differences, unit offsets, or inconsistent time formats in the fields, record the original field content, unit and time format information for field items that fail to match, and generate a field matching offset record table;

[0021] S212: Based on the field items recorded in the field matching offset record table, perform semantic similarity calculation between the original name and the target name of each field, and combine the unit type and time format, perform field label comparison and unit conversion operations through the semantic mapping matrix and conversion rules, calculate and obtain the field comparison conversion balance, and output the converted field name, unit and time standard form according to the conversion matrix corresponding to each field, and establish a standard field conversion mapping table;

[0022] S213: Based on the field items in the standard field conversion mapping table, the frequency distribution of each field in different regional datasets by month is statistically analyzed, ambiguous field combinations are identified, a field classification threshold is set to classify the repetitive fields by frequency, and ambiguous fields are uniquely assigned according to the frequency main class, generating a unified field sequence table.

[0023] As a further aspect of the present invention, the calculation formula for the field comparison and conversion balance is as follows:

[0024] ;

[0025] in, Indicates the first The conversion balance of each field in the comparison and conversion. Indicates the first The semantic weight factor for each item participating in the comparison (set between 0.25 and 0.75 based on historical comparison success rates). Indicates the original field in the 1st position. The length of each dimension (e.g., character length, number of characters per unit). This indicates the standard length of the target field in the same dimension. For unit or time format in the first The conversion complexity coefficient of the item is defined as 1 for same unit, 2 for number system conversion, and 3 for semantically ambiguous conversion.

[0026] As a further aspect of the present invention, the step of obtaining the grouped weighted normalization matrix specifically includes:

[0027] S311: Obtain the service usage count, resource configuration quantity, job function distribution, financial expenditure details and institution level identifier fields from the unified field sequence table, and match the above field items according to the same timestamp, aggregate the corresponding values ​​of the five types of fields under the same month into a multi-dimensional data frame, and establish an independent numerical distribution set for each type of field, respectively count the minimum value, maximum value and interval division point, project the original values ​​to the belonging interval according to the field category, and generate a time dimension numerical distribution interval group;

[0028] S312: Based on the resource allocation quantity and job function distribution fields in the time dimension numerical distribution interval group, the resource allocation and job function are proportionalized according to the weight of the fields. The total proportion of the configuration fields is set to one. The normalized value of each field in the resource dimension is calculated and an allocation vector is constructed. The job dimension is independently normalized according to the type of function. The service usage frequency and fiscal expenditure fields are linearly standardized according to the minimum and maximum values ​​in their distribution intervals. After normalization, a multi-field normalized vector matrix is ​​generated.

[0029] S313: Based on the association and matching of all field normalized values ​​in the multi-field normalized vector matrix with the institution level identifier, divide the entire normalized data set according to the level identifier, set the level group code such as level 1, level 2, and level 3, respectively count the set of field dimension values ​​in the normalized vector under each level group, construct the field mean table under the institution level dimension, perform mean superposition operation on the field normalized values ​​of the same group, and establish a grouped weighted normalized matrix.

[0030] As a further aspect of the present invention, the step of obtaining the graphic coordinate mapping parameter table specifically includes:

[0031] S411: Based on the grouped weighted normalization matrix, extract the resource allocation field, service usage field and job structure field respectively, count the total normalization value and the number of field items in each level group, calculate the ratio of the normalized mean of total resource allocation, the normalized mean of service usage and the normalized mean of job under each level, form a proportional factor vector group under the level dimension, and generate a resource efficiency job ratio matrix.

[0032] S412: Based on the grade labels in the resource efficiency job ratio matrix, retrieve the city codes corresponding to each rehabilitation service institution, map the institution codes to the administrative region coordinate dictionary, extract the latitude and longitude coordinates and assign the current grade label and normalized ratio information, construct a coordinate frame dataset, and then add a number to each coordinate record according to the order of the institution list and define it as the starting point for X and Y plane rendering, and generate grade label coordinate rendering sequence frames.

[0033] S413: Based on the coordinate data and normalized ratio vector recorded in the rendering sequence frame of the grade label coordinates, respectively set the resource ratio dimension to the radar chart radius vector, the utilization efficiency ratio dimension to the height of the bar chart coordinate axis, and the job coverage ratio dimension to the heat map color scale index, assign three types of mapping parameters to each set of coordinate points, and establish a graphic coordinate mapping parameter table.

[0034] As a further aspect of the present invention, the steps for obtaining the visualization assessment chart set of rehabilitation services are as follows:

[0035] S511: Based on the organization number field in the graphic coordinate mapping parameter table as the horizontal axis identifier, the resource ratio field is sequentially bound to the radar chart layer, and the radial region of the radar chart is constructed with the normalized value. For each organization node, vertices are generated in the polar coordinate system and closed regions are drawn according to the weights. After completing the graphic rendering, a resource radar rendering layer is generated.

[0036] S512: Based on the sequence of organization nodes in the resource radar rendering layer, synchronously extract the utilization efficiency ratio field from the graphic coordinate mapping parameter table, map the normalized value to the Y-axis coordinate position of the bar chart layer, place the corresponding organization bar height according to the proportion, load the job coverage ratio field, map it to the color level value of the heat map layer, and assign it to the organization coordinate cell corresponding to the map layer to generate a multi-layer visual component set.

[0037] S513: For each layer component in the multi-layer visual component set, set a time dimension scroll bar component for horizontal time node switching, configure a layer switching button to implement the mutually exclusive logic of radar, bar and heat map display, load the comparison item selection control and bind the response action to the main layer, aggregate all component structures into the same interactive interface visual area and establish linkage configuration rules to generate a rehabilitation service visualization assessment chart group.

[0038] A visualization assessment system for rehabilitation services, comprising:

[0039] The indicator system design module is used to execute S1: obtain the dimension indicator entries in the TRIC and FRAME structures and collect data sources, identify the indicator codes, measurement units, and level classifications in the data sources, and set the collection methods and evaluation thresholds in combination with the indicator dimensions to generate a set of rehabilitation dimension indicator numbers;

[0040] The unified data structure module is used to execute S2: based on the rehabilitation dimension indicator number set, obtain the structured questionnaire platform data and reporting database of rehabilitation service institutions in each region, detect the coding method, time format and unit type in each data, perform label comparison and unit conversion operations through field similarity, and generate a unified field sequence table;

[0041] The indicator value integration module is used to execute S3: Based on the service usage frequency, resource allocation quantity, job function distribution, fiscal expenditure details and institution level identifier in the unified field sequence table, obtain the value distribution range under the same time dimension, perform weighted proportional allocation on resource allocation quantity and job function distribution, perform standardized interval normalization on service usage frequency and fiscal expenditure details, and synchronously group all normalization results according to the rehabilitation service institution level identifier to generate a grouped weighted normalization matrix;

[0042] The graphic coordinate generation module is used to execute S4: based on the grouped weighted normalization matrix, it calculates the resource ratio, utilization efficiency ratio and job coverage ratio of each group, extracts the city code and connects it to the administrative region coordinate dictionary, sets the starting point of planar rendering according to the distribution order of rehabilitation service institutions, performs parameter mapping, generates a graphic coordinate mapping parameter table, sets the time dimension scrolling component and comparison item selection control, loads all graphics into the main view area of ​​the interactive interface, and generates a group of visual evaluation charts for rehabilitation services.

[0043] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0044] This invention achieves effective organization and standardized management of scattered data sources by constructing a unified number set covering multi-dimensional indicators. It improves the compatibility and accuracy of data fusion through field similarity comparison and unit conversion. By normalizing and grouping values ​​such as service usage, resource allocation, job distribution, and fiscal expenditure, it ensures that various indicators are weighted and compared horizontally under unified standards. Visual mapping of core parameters such as resource ratio, efficiency ratio, and coverage ratio enhances the intuitive presentation of evaluation results in terms of time series, spatial dimensions, and indicator weights. Interactive components and layer switching enhance the dynamic response and multi-view switching capabilities of the graphical interface, solving the problems of insufficient structure, single result presentation format, and lack of real-time performance in traditional evaluations.

[0045] The system can be promoted across different regions and service institutions, supports comparative analysis and dynamic monitoring, and achieves the goals of policy formulation, resource allocation, and service quality improvement. It can be widely applied to scenarios such as the evaluation of the rehabilitation service system for people with disabilities, monitoring of policy implementation effectiveness, and performance evaluation of institutional operations, and has high promotional value and practicality. Attached Figure Description

[0046] Figure 1 This is a flowchart of the main steps of the present invention;

[0047] Figure 2 This is a flowchart illustrating the process of obtaining the rehabilitation dimension indicator number set for this invention.

[0048] Figure 3 This is a flowchart illustrating the process of obtaining the unified field sequence table in this invention.

[0049] Figure 4 This is a flowchart of the process for obtaining the grouped weighted normalized matrix in this invention;

[0050] Figure 5 This is a flowchart of the process for obtaining the graphic coordinate mapping parameter table of the present invention;

[0051] Figure 6 This is a flowchart illustrating the process of obtaining a visualization assessment chart for rehabilitation services according to the present invention. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0053] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0054] Please see Figure 1 A method for visually assessing rehabilitation services includes the following steps:

[0055] S1: Obtain the indicator items under the five dimensions of "leadership and governance", "financing mechanism", "human resource allocation", "infrastructure conditions" and "service accessibility and quality" in the TRIC and FRAME structures. Collect data from rehabilitation institution system management documents, financial expenditure records, human resource allocation files, facility equipment ledgers and outpatient usage ledgers. Identify the indicator codes, measurement units and level classifications in the data sources. Combine the indicator dimensions to set the collection methods and evaluation thresholds respectively, and generate a set of rehabilitation dimension indicator numbers.

[0056] S2: Based on the rehabilitation dimension indicator number set, obtain the structured questionnaire platform data and reporting database of rehabilitation service institutions in each region, detect the coding method, time format and unit type in each data, perform label comparison and unit conversion through field similarity, classify and judge the ambiguous data fields according to the time distribution frequency, and generate a unified field sequence table.

[0057] S3: Based on the service usage count, resource allocation quantity, job function distribution, fiscal expenditure details and institution level identifier in the unified field sequence table, obtain the numerical distribution range under the same time dimension, perform weighted proportional allocation on the resource allocation quantity and job function distribution, perform standardized interval normalization on the service usage count and fiscal expenditure details, and synchronously group all normalization results according to the rehabilitation service institution level identifier to generate a grouped weighted normalization matrix.

[0058] S4: Based on the grouped weighted normalization matrix, calculate the resource ratio, utilization efficiency ratio and job coverage ratio of each group, extract the city code and connect it to the administrative region coordinate dictionary, set the X and Y plane rendering start points according to the distribution order of rehabilitation service institutions, map the resource ratio to the radar coordinate system, map the utilization efficiency ratio to the bar chart coordinate axis, map the job coverage ratio to the heat map color scale, and generate a graphic coordinate mapping parameter table;

[0059] S5: Based on the graphic coordinate mapping parameter table, select rehabilitation service institutions as the horizontal axis label, render the radar chart shape area according to the resource ratio, set the height of the bar chart according to the utilization efficiency ratio, load the job coverage ratio to set the map heat level, configure the layer switching button to set the time dimension scrolling component and the comparison item selection control, load all graphics into the main view area of ​​the interactive interface, and generate a group of rehabilitation service visualization assessment charts.

[0060] The rehabilitation dimension indicator number set includes indicator number structure, measurement unit system, level classification rules, data collection method parameters, and evaluation threshold settings. The unified field sequence table includes a unified coding label set, a unified time format set, a unified unit conversion table, field semantic classification labels, and field matching confidence scores. The grouped weighted normalization matrix includes a time dimension grouping table, resource allocation weight matrix, job function ratio matrix, service standardization usage value set, fiscal expenditure normalization dataset, and institution level classification labels. The graphic coordinate mapping parameter table includes a radar chart ratio parameter set, a bar chart efficiency parameter set, a heat map coverage parameter set, an administrative coordinate reference table, and a rendering starting point sequence. The rehabilitation service visualization assessment chart set includes a radar chart layer, a bar chart layer, a heat map layer, a layer switching control, a time scrolling component, a comparison selection control, and a main view area display box.

[0061] Please see Figure 2 Step S1 is as follows:

[0062] S111: Obtain five types of data sources: rehabilitation institution system management documents, financial expenditure records, human resource allocation files, facility equipment ledgers, and outpatient usage ledgers. Perform structural parsing on the record fields in each data source, extract the indicator names, coding formats, measurement units, and level classification parameters from the fields, and map them one by one to the dimensions of leadership and governance, financing mechanisms, human resource allocation, infrastructure conditions, and service accessibility and quality according to the field names. Perform deduplication on the coding items with duplicate fields in the mapping relationship to generate a dimension indicator field mapping set.

[0063] Data sources are obtained from rehabilitation institutions, including five categories: institutional management documents, financial expenditure records, human resource allocation files, facility equipment ledgers, and outpatient usage ledgers. Field deconstruction is then performed. Institutional management documents typically contain text documents such as organizational structure descriptions, institutional charters, and regulations. Fields related to the "leadership and governance" dimension need to be extracted using field-level keyword positioning. For example, keywords such as "institutional charter," "decision-making mechanism," and "supervisory body" can be set, and the corresponding field name "governance structure type" can be recorded, with values ​​set as "single decision-making level," "hierarchical control," or "multi-party collaboration." High-frequency structural units of this field can be extracted as primary fields using text quantity analysis. For financial expenditure records, indicators related to "financing mechanism" are extracted based on project classification fields, such as "proportion of fiscal appropriations" and "proportion of project special funds." The code for the "proportion of fiscal appropriations" field is set to ZC001, and the unit of measurement is "percentage." For example, if the fiscal appropriation in 2022 was 1.8 million yuan and the total expenditure was 3 million yuan, then the calculated value for this field would be:

[0064] ;

[0065] The mapping dimension is set to "funding mechanism". The "percentage of rehabilitation professionals" field is extracted from the human resource allocation file, with its field code set to RL002 and the unit of measurement being "percentage". If a rehabilitation institution has 12 registered rehabilitation therapists and a total of 30 employees, the calculated value is:

[0066] ;

[0067] The facility equipment ledger field is set to extract the "Rehabilitation Bed Allocation Rate" field, coded SS004, with the unit being "beds / 100 people". It measures the number of beds allocated per 100 people served. For example, if an institution serves 500 people and has 35 rehabilitation beds, then the indicator value is:

[0068] ;

[0069] The "Outpatient Follow-up Rate" field, coded FQ006, is extracted from the outpatient ledger field. The unit of measurement is percentage. If the number of follow-up visits in a month is 450 and the total number of outpatient visits is 1000, then the follow-up rate is:

[0070] ;

[0071] Next, the above fields are grouped according to the dimensions they point to, forming a five-dimensional indicator dictionary. Fields with the same field code are deduplicated. For example, if "proportion of fiscal appropriations" is mentioned in both fiscal expenditure records and system management documents, the one with more detailed data records is retained as the standard, and redundant parts are removed. Simultaneously, fields are categorized and aggregated by dimension according to their field codes, forming the following indicator field mapping table:

[0072] Table 1 Indicator Field Mapping Table

[0073] Indicator Name Field Encoding Unit of measurement Classification Belonging Dimension Governance Structure Types ZL001 text Three-level classification (levels 1-3) Leadership and Governance Proportion of fiscal appropriations ZC001 % Four levels (<30%, 30~50%, 50~70%, >70%) Fundraising Mechanism Percentage of recovered patients RL002 % Four levels (<20%, 20~40%, 40~60%, >60%) Human resource allocation Bed occupancy rate SS004 Zhang / 100 people Three levels (<5, 5~8, >8) Infrastructure conditions Outpatient return visit rate FQ006 % Three tiers (<30%, 30~60%, >60%) Service accessibility and quality

[0074] As shown in Table 1, the structure of each type of field has been parsed and deduplicated, and then grouped and aggregated by dimension to obtain the final set of dimension indicator field mappings.

[0075] S112: Based on the dimension indicator field mapping set, extract the measurement unit and level classification parameters of each indicator item, establish an indicator collection structure template, and divide the data source categories of the fields into three collection methods: document collection, structured record extraction, and ledger statistics. Match the collection method with the corresponding field structure template one by one, and attach the collection method to the indicator code item to generate an indicator collection structure combination table.

[0076] Based on the dimensional indicator field mapping set, each indicator item in the record is read, and the corresponding measurement unit and level classification parameters are determined. First, it is necessary to determine the original data source format of each field. Institutional management documents are text files, and are collected using keyword extraction and structured processing (converting the extracted keywords and their context into a standardized 'field-value' pair format), categorized as "document collection type." For example, the "governance structure type" field needs to be extracted using keywords such as "management committee," "executive committee," and "supervision level" in the paragraph, and the field type is marked as ZL001. Fiscal expenditure records and personnel files are generally structured tables; the fields "proportion of fiscal appropriations" and "percentage of recovered personnel" are set as "structured record extraction type," using direct field name matching combined with data row and column coordinates for extraction. The data collection method involves various approaches. For example, the "Source of Funds" and "Amount" columns in the table form a horizontal parsing logic. The fields "Outpatient Follow-up Rate" and "Bed Allocation Rate" involve long-term time-series data collection and statistics, so they are classified as "Ledger Statistics". Their values ​​need to be calculated by accumulating the number of visits or facilities over time. The collection window is set to a monthly unit to achieve periodic data statistics. Furthermore, the collection method is bound to the field name one by one. For example, the collection method of field ZC001 is bound to "Structured Record Extraction", and the collection method of field FQ006 is bound to "Ledger Statistics". Finally, all collection method information, field codes, and their original field names are concatenated to form the collection structure combination information, which represents the uniqueness of each indicator and its structural processing method, and finally generates an indicator collection structure combination table.

[0077] S113: Based on the collection method and field level classification parameters of each indicator in the indicator collection structure combination table, read the measurement unit and level classification value range pointed to by each indicator, compare and calculate the level parameter value with the reference level threshold, mark the judgment threshold parameter corresponding to each level interval, aggregate and configure the indicator code with the collection method and judgment threshold, and generate a set of rehabilitation dimension indicator numbers.

[0078] Based on the indicator collection structure combination table, the collection method information for each indicator field is statistically analyzed. The level classification parameters attached to the fields are extracted. The range values ​​for each level need to be extracted and compared with the reference benchmark value to determine the appropriate threshold parameter for each level. For example, the "Proportion of Fiscal Appropriations" field is set to four levels: below 30% is Level 1, 30%~50% is Level 2, 50%~70% is Level 3, and above 70% is Level 4. The reference benchmark value is 50%. If an institution's value for this item is 45%, it is classified as Level 2. The level parameters and collection methods are then aggregated and configured to further form the field and its... The mapping relationship between the thresholds of the corresponding levels is as follows: For example, the level parameter distribution of the "RL002" field is set to <20%, 20%~40%, 40%~60%, and >60%. The collection method of this field is structured record extraction, and the reference benchmark value is set to 40%. If the proportion of registered rehabilitation personnel in the institution is 38%, then the level is Level 2. At the same time, the field collection period is set to quarterly. Then, the field collection method, collection period and level judgment threshold are combined as the indicator aggregation configuration item. Finally, the indicator number configuration information table shown in the table below is constructed to generate the rehabilitation dimension indicator number set.

[0079] Table 2. Rehabilitation Dimension Indicator Numbering Configuration Table

[0080] Indicator coding Data collection method Threshold setting Grade classification range Collection cycle ZC001 Structured record extraction class Reference benchmark value: 50%; Grade judgment value: 45%. <30%，30~50%，50~70%，>70% Monthly RL002 Structured record extraction class Reference benchmark value: 40%; Grade judgment value: 38%. <20%，20~40%，40~60%，>60% Quarter SS004 Ledger Statistics Reference benchmark: 6 sheets per 100 people; Judgment benchmark: 7 sheets per 100 people <5，5~8，>8 Half a year

[0081] As shown in Table 2, a complete field collection structure, judgment threshold and level range configuration have been formed, and this information has been aggregated to form a number set, which finally yields the rehabilitation dimension indicator number set.

[0082] Please see Figure 3 Step S2 is as follows:

[0083] S211: Obtain structured questionnaire platform data and reporting database of rehabilitation service institutions in various regions, extract field content, field code and field unit in each dataset, cross-match field nouns with indicator field names in the rehabilitation dimension indicator number set, identify whether there are naming differences, unit offsets, or inconsistent time formats in the fields, record the original field content, unit and time format information for field items that fail to match, and generate a field matching offset record table;

[0084] Combining structured questionnaire platform data and reported databases from rehabilitation service institutions in various regions, the original data field content is extracted and broken down into four basic parameters: field name, field code, time annotation format, and unit of measurement. A one-to-one comparison is performed between the field name and the standard field items centrally registered with rehabilitation dimension indicator numbers. The comparison operation employs a dual verification mechanism of string similarity and semantic similarity (setting a direct match when the string edit distance is less than or equal to 2, or a field pending judgment when the edit distance is greater than 2 but the semantic keyword hit rate reaches 100%). In the field string similarity calculation, a string edit distance not exceeding 2 is set as an approximate matching condition. If the field "rehabilitation bed configuration rate" is labeled as "bed occupancy density" or "rehabilitation bed density," its edit distances are 5 and 4 respectively, exceeding the set upper limit. However, through semantic keyword matching, the two keywords "bed" and "rehabilitation" are extracted. If the core keyword hit rate reaches 100%, it is determined to be a field to be judged and needs to be transferred to the unit conversion stage for further analysis. The unit conversion operation performs unit standardization calculations for fields with inconsistent measurement units. For example, if one field's unit is "number of beds / 100 people" and another is "sheets / thousand people", the units need to be converted to a unified form. The conversion benchmark is set to "sheets / hundred people". Then, "sheets / thousand people" needs to be divided by 10 for conversion. If the original field value is 25 sheets / thousand people, it will be converted to 2.5 sheets / hundred people. The time format is uniformly set to "YYYY-MM-DD". If the original data uses the "year / day" format (such as 2024 / 12), the month information needs to be supplemented and the separator needs to be unified. Fields whose belonging cannot be determined by the three rules of semantic, unit and time matching need to be recorded as suspected fields. Their original tag name, unit type and time format records are collected and summarized into the offset record table to form a field matching offset record table.

[0085] S212: Based on the field entries in the field matching offset record table, perform semantic similarity calculation between the original name and the target name of each field. Combined with unit type and time format, execute field label comparison and unit conversion operations through a semantic mapping matrix and conversion rules, using the following formula:

[0086] ;

[0087] The calculation obtains the field comparison and conversion balance value, and outputs the converted field name, unit, and time standard form based on the corresponding conversion matrix of each field, establishing a standard field conversion mapping table; among which... Indicates the first The conversion balance of each field in the comparison and conversion. Indicates the first The semantic weight factor for each item participating in the comparison (set between 0.25 and 0.75 based on historical comparison success rates). Indicates the original field in the 1st position. The length of each dimension (e.g., character length, number of characters per unit). This indicates the standard length of the target field in the same dimension. For unit or time format in the first The conversion complexity coefficient of the item is defined as 1 for the same unit, 2 for number system conversion, and 3 for semantically ambiguous conversion.

[0088] Based on the fields registered in the field matching offset record table, the original label, target label, original unit, standard unit, and time format of each field are extracted to construct a comparison vector set. The label similarity is converted into vector distance using a dimensional weighted method. The unit conversion item introduces a semantic deviation correction factor and combines it with the time format normalization index. A comprehensive conversion formula is set, and the field semantic matching and measurement unit conversion operations are performed. The calculation is performed using the formula.

[0089] Taking the label comparison of the field "Rehabilitation Service Resources per Thousand People" with the standard field "Rehabilitation Bed Allocation Rate" as an example, let's assume the original label character length is... Standard label length Calculate the complexity coefficient (Since the unit needs to be multiplied by a conversion factor of 10), set semantic weights. The corresponding single-item calculation value is:

[0090] ;

[0091] If this calculation is embedded into all dimensions, the resulting field will be... Conversion index value Based on experience, the threshold was set to 2.5 when comparing score ranges. If a field can be grouped into a target standard field, perform field standardization and replacement, and record the corresponding relationship between the fields before and after conversion, the conversion method, and the conversion factor in the table. For example, the original value of the field "number of beds / thousand people" is 25 beds. After conversion, the unit is "beds / hundred people", the conversion value is 2.5 beds, and the conversion factor is 0.1. The original time format is "2023 / 04", which is uniformly adjusted to "2023-04-01". Record this in the conversion field table, as follows:

[0092] Table 3 Field Conversion Mapping Table

[0093] Original field label Standard field name Original unit Standard units Time format Conversion factor Character difference score Number of beds per thousand people Rehabilitation bed allocation rate Zhang / Thousands of people Zhang / 100 people 2023 / 04 0.1 1.697 Rehabilitation service resources per thousand people Rehabilitation bed allocation rate Zhang / Thousands of people Zhang / 100 people April 2023 0.1 1.697

[0094] As shown in Table 3, standard field mapping items were established through field semantic comparison and conversion operations, and the field comparison and conversion balance was obtained through calculation. Based on the comparison results between its value and the threshold, a standard mapping is performed to generate a standard field conversion mapping table;

[0095] The formula's operational logic primarily measures the strength of differences between the field to be matched and the standard field across multiple dimensions, thereby determining whether they possess the potential for a unified mapping after normalization. Firstly, the formula employs a difference normalization method, that is, by normalizing a certain dimension of the original field... Standard attributes corresponding to the target field The difference between them is calculated to represent the numerical offset between them in that dimension, and the square root operation is used. This is to convert the complexity coefficient. To mitigate the difficulty of conversion, nonlinear adjustments are made by placing the fraction in the denominator and taking the square root; secondly, multiplication terms are used. It introduces semantic weights This factor measures the importance of the current dimension in the overall judgment, reflecting the uneven impact of differences in semantics, units, time, and other dimensions. For example, the semantic importance of field labels is usually higher than the difficulty of unit conversion, so they can be given higher weight. Furthermore, the overall calculation uses absolute values. This is to avoid positive and negative differences canceling each other out, thus ensuring that the offsets in all dimensions are fully accounted for in the total; finally, the contribution values ​​of the differences under all dimensions are summed and then divided by the number of dimensions. Averaging is performed to obtain the standardized field conversion balance. This value serves as the basis for merging judgments. The lower the value, the smaller the difference between the field and the standard field in each dimension, and the more substitutable it is. The overall logic reflects a composite measurement mechanism that harmonizes field similarity across multiple dimensions and penalizes complexity, ensuring that the generated comparison results are highly robust and operable.

[0096] The field comparison and conversion equilibrium value is a composite indicator used to measure the degree of difference between the original data field and the standard field under multi-dimensional attributes. It reflects the overall similarity and conversion cost between the two in terms of label semantics, unit format, and time format. This equilibrium value comprehensively considers the degree of semantic matching, unit conversion complexity, and time expression differences. After weighting and normalizing these differences, a quantitative indicator is formed. The smaller the value, the closer the structure, meaning, and expression of the original field and the standard field are, and the more directly it can be included in the unified field system. Conversely, if the value is large, it indicates that there is a large offset or ambiguity between the fields, which cannot be directly mapped and requires further analysis or elimination. Therefore, the field comparison and conversion equilibrium value is not only the core basis for determining field ownership in the standardization process, but also reflects the compatibility and mapping stability of fields in structured integration, and is a key control parameter in the field merging process.

[0097] S213: Based on the field items in the standard field conversion mapping table, the frequency distribution of each field in different regional datasets by month is statistically analyzed, ambiguous field combinations are identified, and repeated fields that appear more than 20% of the total monthly samples are classified by category frequency. The field classification threshold is set to monthly repetition frequency > 5 times, and ambiguous fields are uniquely assigned according to the frequency main class, generating a unified field sequence table.

[0098] Based on the field names and unified units in the standard field conversion mapping table, the frequency of the field in different regional datasets is statistically analyzed monthly. The frequency of the same field in multiple datasets is merged by month, and the field classification type is determined based on the cumulative frequency. Fields that have appeared more than 6 months in total and have repeated more than 5 times per month are set as primary category fields, and the category identifier of the field is recorded. The classification threshold is set as "monthly repetition frequency > 5". If the field "rehabilitation bed allocation rate" is repeated more than 6 times per month in 9 of the 12 regions, it is classified as a high-frequency primary category field. Ambiguous labels under the field, such as "rehabilitation bed ratio" and "bed supply rate", are uniformly assigned to the field category, a unique classification code is established, the field merging configuration is completed, and a unified field sequence table is established.

[0099] Please see Figure 4 Step S3 is as follows:

[0100] S311: Obtain the service usage count, resource configuration quantity, job function distribution, financial expenditure details and institution level identifier fields from the unified field sequence table, match the above field items according to the same timestamp, aggregate the corresponding values ​​of the five types of fields under the same month into a multi-dimensional data frame, and establish an independent numerical distribution set for each type of field, respectively count the minimum value, maximum value and interval division point, project the original values ​​to the belonging interval according to the field category, and generate time dimension numerical distribution interval group;

[0101] Five types of field data were obtained from the unified field sequence table. The values ​​of service usage frequency, resource allocation quantity, job function distribution, fiscal expenditure details, and institution level identifier were broken down one by one, and the corresponding timestamps were extracted. Using the month as the smallest time unit, a joint data frame of the five field sets was constructed in the time dimension. Each field was horizontally aggregated according to the same month, and the minimum, maximum, and dispersion of data distribution for each type of field in that month were recorded. For example, the service usage frequency field, such as "outpatient follow-up visits," ranged from 100 to 600 visits in March 2024; the resource allocation field, such as "number of rehabilitation beds," ranged from 30 to 90 beds; the job function field, such as "number of treatment positions," ranged from 4 to 12 people; and the fiscal expenditure field, such as "equipment expenditure amount," ranged from 50,000 to 250,000 yuan. When establishing the numerical intervals, a five-equal division strategy was adopted, dividing the numerical distribution interval of each field into five equally spaced level intervals. The starting and ending values ​​of the intervals were recorded as reference templates for the normalization processing of that field, forming the interval definition table shown below:

[0102] Table 4. Numerical Distribution Intervals

[0103] Field Name Time dimension Minimum value Maximum value Interval partition number Span of each interval Number of outpatient follow-up visits 2024-03 100 600 5 100 Number of rehabilitation beds 2024-03 30 90 5 12 Number of treatment positions 2024-03 4 12 5 1.6 Equipment expenditure amount 2024-03 50000 250000 5 40000

[0104] As shown in Table 4, the value range of each field is uniformly divided into 5 intervals according to the time dimension. The normalization and weighting of various types of data in the future need to be based on the discrete interval range set in this table to generate the time dimension numerical distribution interval group.

[0105] S312: Based on the resource allocation quantity and job function distribution fields in the time dimension numerical distribution interval group, the resource allocation and job function are proportionalized according to the weight of the fields. The total proportion of the configuration fields is set to one. The normalized value of each field in the resource dimension is calculated and an allocation vector is constructed. The job dimension is divided into independent normalization according to the function type. The service usage frequency and fiscal expenditure fields are linearly standardized according to the minimum and maximum values ​​in their distribution intervals. After normalization, a multi-field normalized vector matrix is ​​generated.

[0106] Based on the time-dimensional numerical distribution intervals, the resource allocation quantity field and the job function distribution field are extracted. Numerical configuration items such as "number of rehabilitation beds" and "number of rehabilitation equipment" are summed to form the total configuration base. Then, an allocation coefficient is extracted for each configuration field according to its proportion. For example, in an institution, if the number of rehabilitation beds is 60 and the number of equipment is 90, the total configuration is 150, corresponding to configuration proportions of 0.4 and 0.6, which are used as weighting factors in the subsequent normalization value construction. The job function fields "number of treatment positions" and "number of management positions" are set to 8 and 2 respectively, and normalized according to job category proportions, so the treatment proportion is 0.8 and the management proportion is 0.2. At that time, the service usage frequency field "re-visit visits" was 500, and its corresponding field range was 100 to 600. Linear normalization was applied, and the normalized value was calculated as (500-100) / (600-100)=0.8. The fiscal expenditure field "equipment expenditure amount" was 210,000 yuan, and the corresponding range was 50,000 to 250,000. Then the normalized value was (210,000-50,000) / (250,000-50,000)=0.8. Finally, the configuration ratio vector [0.4, 0.6], the job normalization vector [0.8, 0.2] and the service and expenditure normalization values ​​were concatenated to form a unified field matrix, generating a multi-field normalized vector matrix.

[0107] S313: Based on the association and matching of all field normalized values ​​in the multi-field normalized vector matrix with the organization level identifier, divide the entire normalized data set according to the level identifier, set the level group code such as level 1, level 2, level 3, respectively, and count the set of field dimension values ​​in the normalized vector under each level group, construct the field mean table under the organization level dimension, perform mean superposition operation on the field normalized values ​​of the same group, and establish a grouped weighted normalized matrix.

[0108] Based on the normalized values ​​of all fields in the multi-field normalized vector matrix, an association and merging operation is performed with the institution level identifier field. The institution level identifier is set into three categories: "Level 1", "Level 2", and "Level 3". Each normalized vector is labeled according to its institution level and grouped and aggregated. Within the same level group, the normalized value set of all fields corresponding to the normalized vectors is extracted. The mean of the normalized value of each field within the group is calculated to obtain the weighted center of the field for that level. For example, in a "Level 1" institution, the "normalized value of repeat visits" is 0.6, 0.7, and 0.8, so the mean of this field in the Level 1 group is 0.7. The job function normalized vector is extracted by extracting the average value within each group according to the field dimension. The configuration fields are weighted by the original weighting coefficient and then accumulated to construct a grouped configuration total indicator matrix. Finally, a grouped mean summary table is constructed by level and output as follows:

[0109] Table 5. Normalized Mean of Grouping Field

[0110] Level label Standardize outpatient follow-up values Normalized value of fiscal expenditure Configure vector weighted mean Job Vector Normalized Mean Level 1 0.70 0.65 0.58 0.72 Level 2 0.55 0.50 0.42 0.60 Level 3 0.40 0.30 0.35 0.50

[0111] As shown in Table 5, the field normalized values ​​are effectively grouped in the rank dimension, and the matrix structure has been configured according to the association between field attributes and ranks, thus establishing a grouped weighted normalized matrix.

[0112] Please see Figure 5 Step S4 is as follows:

[0113] S411: Based on the grouped weighted normalized matrix, extract the resource allocation field, service usage field and job structure field respectively, count the total normalized value and the number of field items of each field in each level group, calculate the ratio of the normalized mean of total resource allocation, the normalized mean of service usage and the normalized mean of job under each level, form a proportional factor vector group under the level dimension, and generate a resource efficiency job ratio matrix.

[0114] Based on the resource, service usage, and job structure fields stored in the grouped weighted normalization matrix, normalization value sets corresponding to level 1, level 2, and level 3 institutions are extracted respectively. The sum of normalization values ​​and the number of fields are calculated for each field item within each group. For resource configuration fields such as "rehabilitation bed allocation rate" and "equipment density," the average value within the group is calculated. If the corresponding normalization values ​​in a certain level group are 0.60, 0.72, and 0.68 respectively, then the average resource value is (0.60 + 0.72 + 0.68) / 3 = 0.6667. For service usage fields such as "outpatient visits normalization value," the value is 0.75. The average values ​​for the three categories, 0.85 and 0.70, are 0.7667. For job-related fields such as "treatment job coverage," the average values ​​are 0.80, 0.85, and 0.90, with a final average of 0.85. Then, for each group, the three average values ​​are normalized using the formulas: Resource Ratio = Average Resource / Total, Efficiency Ratio = Average Service / Total, and Job Ratio = Average Job Ratio / Total. Assuming the sum of the three average values ​​for this group is 2.2834, then the Resource Ratio = 0.6667 / 2.2834 ≈ 0.292, Efficiency Ratio ≈ 0.336, and Job Ratio ≈ 0.372. This process is repeated for all level groups, generating the ratio matrix shown below.

[0115] Table 6 Resource Efficiency Ratio Matrix for Each Level Group

[0116] grade Resource normalized mean Use normalized mean Normalized average of positions resource ratio efficiency ratio Job ratio Level 1 0.6667 0.7667 0.8500 0.292 0.336 0.372 Level 2 0.5400 0.6200 0.7100 0.295 0.339 0.366 Level 3 0.4300 0.5200 0.6200 0.290 0.351 0.359

[0117] As shown in Table 6, the resource efficiency job ratio matrix has been obtained by normalizing the mean and performing total allocation calculations, outputting the ratio vector by level.

[0118] S412: Based on the grade labels in the resource efficiency job ratio matrix, retrieve the corresponding city codes of each rehabilitation service institution, map the institution codes to the administrative region coordinate dictionary, extract the latitude and longitude coordinates and assign the current grade label and normalized ratio information, construct the coordinate frame dataset, and then add a number to each coordinate record according to the order of the institution list and define it as the starting point for X and Y plane rendering, and generate the grade label coordinate rendering sequence frame.

[0119] Based on the resource efficiency job ratio matrix, using the grade labels "Level 1", "Level 2", and "Level 3" as primary keys, the "Institution Code" field in the original form of rehabilitation institutions is retrieved, and a one-to-one mapping between institutions and grades is established. The institution code is used as a connection field, and latitude and longitude positioning is performed by referring to the administrative region coordinate dictionary. Assuming that the institution with institution code 110001 matches the coordinates (116.4074, 39.9042) in the dictionary, and the institution's grade is "Level 1", then the grade and coordinates are used to form a triplet data frame [110001, Level 1, (116.4074, 39.9042)]. All institution records are numbered in ascending order according to the institution code. Let their rendering order be the institution list sequence number 1~N. The coordinates of the X and Y planes are used as the initial point coordinates for drawing according to the number. If 110001 is the 4th institution, then the corresponding rendering starting point is X=4, Y=4. After repeating the processing, a rendering sequence frame containing the institution number, grade label, coordinate latitude and longitude, and drawing starting point is constructed, resulting in the grade label coordinate rendering sequence frame.

[0120] S413: Based on the coordinate data and normalized ratio vector recorded in the rendering sequence frame of the grade label coordinate, the resource ratio dimension is mapped to the radar chart radius vector, the efficiency ratio dimension is mapped to the height of the bar chart coordinate axis, and the job coverage ratio dimension is mapped to the heat map color scale index. Three types of mapping parameters are assigned to each set of coordinate points, and a graphic coordinate mapping parameter table is established.

[0121] Based on the numbered coordinate records in the rendering sequence frames of the grade label coordinates, the resource ratio field value is mapped to the radial axis value of the radar chart. Assuming a resource ratio of 0.292, the radial vector length is set to the base radius multiplied by 0.292. For example, if the base radius is 100, the current radius is 29.2. The efficiency ratio of 0.336 is mapped to the Y-axis height of the bar chart, resulting in a height of 33.6 units at a uniform scale. The job ratio of 0.372 is mapped to the color value of the heatmap. Assuming the heatmap color level is in the range of 0-255, the corresponding color value is 0.372 × 255 ≈ 95. The mapping results of these three layers are bound one-to-one with the coordinate points, establishing a mapping relationship between coordinate values ​​and layer parameters. Finally, all rendering control parameters required by the visible layers and the coordinate point codes are synchronously summarized and output to establish a graphic coordinate mapping parameter table.

[0122] Please see Figure 6The S5 steps are as follows:

[0123] S511: Based on the organization number field in the graphic coordinate mapping parameter table as the horizontal axis identifier, the resource ratio field is sequentially bound to the radar chart layer, and the radial region of the radar chart is constructed with the normalized value. For each organization node, vertices are generated in the polar coordinate system and closed regions are drawn according to the weights. After completing the graphic rendering, a resource radar rendering layer is generated.

[0124] Based on the "Institution Number" field in the graphical coordinate mapping parameter table as the horizontal axis, rehabilitation service institutions are arranged in ascending order. The normalized value corresponding to the resource ratio field is set as the vector magnitude value under the polar coordinates of the radar chart. The angle axis of the radar chart is divided into 6 dimensions according to the number of fields, corresponding to indicators such as bed density, equipment ratio, and staff load. For each institution, the normalized value is extracted sequentially, such as [0.25, 0.40, 0.50, 0.30, 0.35, 0.45]. Assuming the base radius is 120 units, the radial lengths of each point are 30, 48, 60, 36, 42, and 54. The six points are connected in sequence to form a closed hexagonal area, and the edges are filled with gray with an opacity of 0.5, which constitutes the resource shape area of ​​the radar chart. After all institutions are rendered in sequence, they are merged into the main layer to generate a resource radar rendering layer.

[0125] S512: Based on the sequence of organizational nodes in the resource radar rendering layer, synchronously extract the utilization efficiency ratio field from the graphic coordinate mapping parameter table, map the normalized value to the Y-axis coordinate position of the bar chart layer, place the corresponding organizational bar heights proportionally, load the job coverage ratio field, map it to the color level value of the heat map layer, and assign it to the corresponding organizational coordinate cell of the map layer to generate a multi-layer visual component set.

[0126] Based on the organizational number order in the resource radar rendering layer, the "Utilization Efficiency Ratio" field is obtained and its normalized value is mapped to the Y-axis height of the bar chart. The drawing scale is set to the normalized value × 100 units. If the efficiency ratio of organizational A is 0.76, the corresponding bar height is 76 units. At the same time, the "Job Coverage Ratio" field is extracted from the parameter table and mapped to the color level value of the map heatmap. The color level range is 0 to 255. If the job coverage ratio is 0.52, the color level value is 133. The map grid color is assigned to RGB (255, 122, 122). The bar charts and heatmaps of all organizations are drawn on their respective layers and bound to the organizational numbers for subsequent linkage. Finally, the bar charts and heatmaps are merged into the rendering set to generate a multi-layer visual component set.

[0127] S513: For multi-layer visual components, set up a time dimension scroll bar component for horizontal time node switching, configure layer switching buttons to implement mutually exclusive logic for radar, bar and heat map display, load comparison item selection control and bind response action to the main layer, aggregate all component structures into the same interactive interface visual area and establish linkage configuration rules to generate a group of rehabilitation service visualization assessment charts.

[0128] For the radar layer, bar layer, and heat map layer already established in the multi-layer visual component set, a horizontal scroll bar component is set for time series display. The component supports monthly scrolling, with an initial time step of 1 month. A button-type layer switching control is loaded, and the three types of graphics are set to mutually exclusive visibility states. The button trigger logic is "selecting a layer activates it while hiding other layers". A comparison item selection control is set for horizontal comparison between institutions. The selection box item is sourced from the institution number field. The control linkage logic is that changes in the option trigger a refresh of all layers, which are synchronously mapped to the positioning point in the view area. All layers and components are merged into an interactive layout and loaded into the front-end main view frame to generate a set of rehabilitation service visualization assessment charts.

[0129] A visualization assessment system for rehabilitation services, comprising:

[0130] The indicator system design module is used to execute S1: obtain the dimension indicator entries in the TRIC and FRAME structures and collect data sources, identify the indicator codes, measurement units, and level classifications in the data sources, and set the collection methods and evaluation thresholds in combination with the indicator dimensions to generate a set of rehabilitation dimension indicator numbers;

[0131] The unified data structure module is used to execute S2: based on the rehabilitation dimension indicator number set, it obtains the structured questionnaire platform data and reporting database of rehabilitation service institutions in various regions, detects the coding method, time format and unit type in each data, performs label comparison and unit conversion operations through field similarity, and generates a unified field sequence table;

[0132] The indicator value integration module is used to execute S3: Based on the service usage frequency, resource allocation quantity, job function distribution, financial expenditure details and institution level identifier in the unified field sequence table, it obtains the value distribution range under the same time dimension, performs weighted proportional allocation on resource allocation quantity and job function distribution, performs standardized interval normalization on service usage frequency and financial expenditure details, and synchronously groups all normalization results according to the rehabilitation service institution level identifier to generate a grouped weighted normalization matrix.

[0133] The graphic coordinate generation module is used to execute S4: based on the grouped weighted normalization matrix, it calculates the resource ratio, utilization efficiency ratio and job coverage ratio of each group, extracts the city code and connects it to the administrative region coordinate dictionary, sets the starting point of plane rendering according to the distribution order of rehabilitation service institutions, performs parameter mapping, and generates a graphic coordinate mapping parameter table.

[0134] The visualization module is used to execute S5: based on the graphic coordinate mapping parameter table, it sets the time dimension scrolling component and the comparison item selection control, loads all graphics into the main view area of ​​the interactive interface, and generates a set of visualization assessment charts for rehabilitation services.

[0135] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for visually assessing rehabilitation services, characterized in that, Includes the following steps: S1: Obtain the dimension indicator entries in the TRIC and FRAME structures and collect the data sources. Identify the indicator codes, measurement units, and level classifications in the data sources. Combine the indicator dimensions to set the collection methods and evaluation thresholds respectively, and generate a set of rehabilitation dimension indicator numbers. S2: Based on the rehabilitation dimension indicator number set, obtain the structured questionnaire platform data and reporting database of rehabilitation service institutions in each region, detect the coding method, time format and unit type in each data, perform label comparison and unit conversion operations through field similarity, and generate a unified field sequence table; S3: Based on the service usage count, resource allocation quantity, job function distribution, fiscal expenditure details and institution level identifier in the unified field sequence table, obtain the numerical distribution range under the same time dimension, perform weighted proportional allocation on the resource allocation quantity and job function distribution, perform standardized interval normalization on the service usage count and fiscal expenditure details, and synchronously group all normalization results according to the rehabilitation service institution level identifier to generate a grouped weighted normalization matrix. S4: Based on the grouped weighted normalization matrix, calculate the resource ratio, utilization efficiency ratio and job coverage ratio of each group, extract the city code and connect it to the administrative region coordinate dictionary, set the starting point of plane rendering according to the distribution order of rehabilitation service institutions, perform parameter mapping, and generate a graphic coordinate mapping parameter table. S5: Based on the graphic coordinate mapping parameter table, set the time dimension scrolling component and the comparison item selection control, load all graphics into the main view area of ​​the interactive interface, and generate a group of visual assessment charts for rehabilitation services. The specific steps for obtaining the unified field sequence table are as follows: S211: Obtain structured questionnaire platform data and reporting database of rehabilitation service institutions in various regions, extract field content, field code and field unit in each dataset, cross-match field nouns with indicator field names in the rehabilitation dimension indicator number set, identify whether there are naming differences, unit offsets, or inconsistent time formats in the fields, record the original field content, unit and time format information for field items that fail to match, and generate a field matching offset record table; S212: Based on the field items recorded in the field matching offset record table, perform semantic similarity calculation between the original name and the target name of each field, and combine the unit type and time format, perform field label comparison and unit conversion operations through the semantic mapping matrix and conversion rules, calculate and obtain the field comparison conversion balance, and output the converted field name, unit and time standard form according to the conversion matrix corresponding to each field, and establish a standard field conversion mapping table; S213: Based on the field items in the standard field conversion mapping table, the frequency distribution of each field in different regional datasets by month dimension is statistically analyzed, ambiguous field combinations are identified, a field classification threshold is set to classify the repetitive fields by frequency, and ambiguous fields are uniquely assigned according to the frequency main class to generate a unified field sequence table. The formula for calculating the field comparison and conversion balance is as follows: ； in, Indicates the first The conversion balance of each field in the comparison and conversion. Indicates the first Semantic weight factors for items participating in the comparison Indicates the original field in the 1st position. The length of each dimension This indicates the standard length of the target field in the same dimension. For unit or time format in the first The conversion complexity coefficient of the item.

2. The method for visually assessing rehabilitation services according to claim 1, characterized in that, The specific indicators mentioned above refer to leadership and governance, fundraising mechanisms, human resource allocation, infrastructure conditions, and service accessibility and quality. The data sources include rehabilitation institution management documents, financial expenditure records, human resource allocation files, facility allocation ledgers, and outpatient usage ledgers; Specifically, the execution parameter mapping refers to mapping the resource ratio to the radar coordinate system, mapping the utilization efficiency ratio to the bar chart coordinate axis, and mapping the job coverage ratio to the heat map color scale. The rehabilitation dimension indicator number set includes indicator number structure, measurement unit system, level classification rules, data collection method parameters, and evaluation threshold settings. The unified field sequence table includes a unified coding label set, a unified time format set, a unified unit conversion table, field semantic classification labels, and field matching confidence scores. The grouped weighted normalization matrix includes a time dimension grouping table, a resource allocation weight matrix, a job function ratio matrix, a service standardization usage value set, a fiscal expenditure normalization dataset, and an institution level classification label. The graphic coordinate mapping parameter table includes a radar chart ratio parameter set, a bar chart efficiency parameter set, a heat map coverage parameter set, an administrative coordinate reference table, and a rendering starting point sequence.

3. The method for visually assessing rehabilitation services according to claim 1, characterized in that, The specific steps for obtaining the set of rehabilitation dimension indicator numbers are as follows: S111: Obtain five types of data sources: rehabilitation institution system management documents, financial expenditure records, human resource allocation files, facility equipment ledgers, and outpatient usage ledgers. Perform structural parsing on the record fields in each data source, extract the indicator names, coding formats, measurement units, and level classification parameters from the fields, and map them one by one to the dimensions of leadership and governance, financing mechanisms, human resource allocation, infrastructure conditions, and service accessibility and quality according to the field names. Perform deduplication on the coding items with duplicate fields in the mapping relationship to generate a dimension indicator field mapping set. S112: Based on the dimension indicator field mapping set, extract the measurement unit and level classification parameters of each indicator item, establish an indicator collection structure template, and divide the data source categories of the fields into three collection methods: document collection, structured record extraction, and ledger statistics. Match the collection methods with the corresponding field structure templates one by one, and attach the collection methods to the indicator coding items to generate an indicator collection structure combination table. S113: Based on the collection method and field level classification parameters of each indicator in the indicator collection structure combination table, read the measurement unit and level classification value range pointed to by each indicator, compare and calculate the level parameter value with the reference level threshold, mark the judgment threshold parameter corresponding to each level interval, aggregate and configure the indicator code, collection method and judgment threshold, and generate a set of rehabilitation dimension indicator numbers.

4. The method for visually assessing rehabilitation services according to claim 1, characterized in that, The specific steps for obtaining the grouped weighted normalized matrix are as follows: S311: Obtain the service usage count, resource configuration quantity, job function distribution, financial expenditure details and institution level identifier fields from the unified field sequence table, and match the above field items according to the same timestamp, aggregate the corresponding values ​​of the five types of fields under the same month into a multi-dimensional data frame, and establish an independent numerical distribution set for each type of field, respectively count the minimum value, maximum value and interval division point, project the original values ​​to the belonging interval according to the field category, and generate a time dimension numerical distribution interval group; S312: Based on the resource allocation quantity and job function distribution fields in the time dimension numerical distribution interval group, the resource allocation and job function are proportionalized according to the weight of the fields. The total proportion of the configuration fields is set to one. The normalized value of each field in the resource dimension is calculated and an allocation vector is constructed. The job dimension is independently normalized according to the type of function. The service usage frequency and fiscal expenditure fields are linearly standardized according to the minimum and maximum values ​​in their distribution intervals. After normalization, a multi-field normalized vector matrix is ​​generated. S313: Based on the association and matching of all field normalized values ​​in the multi-field normalized vector matrix with the institution level identifier, divide the entire normalized data set according to the level identifier, set the level group code such as level 1, level 2, and level 3, respectively count the set of field dimension values ​​in the normalized vector under each level group, construct the field mean table under the institution level dimension, perform mean superposition operation on the field normalized values ​​of the same group, and establish a grouped weighted normalized matrix.

5. The method for visually assessing rehabilitation services according to claim 1, characterized in that, The specific steps for obtaining the graphic coordinate mapping parameter table are as follows: S411: Based on the grouped weighted normalization matrix, extract the resource allocation field, service usage field and job structure field respectively, count the total normalization value and the number of field items in each level group, calculate the ratio of the normalized mean of total resource allocation, the normalized mean of service usage and the normalized mean of job under each level, form a proportional factor vector group under the level dimension, and generate a resource efficiency job ratio matrix. S412: Based on the grade labels in the resource efficiency job ratio matrix, retrieve the city codes corresponding to each rehabilitation service institution, map the institution codes to the administrative region coordinate dictionary, extract the latitude and longitude coordinates and assign the current grade label and normalized ratio information, construct a coordinate frame dataset, and then add a number to each coordinate record according to the order of the institution list and define it as the starting point for X and Y plane rendering, and generate grade label coordinate rendering sequence frames. S413: Based on the coordinate data and normalized ratio vector recorded in the rendering sequence frame of the grade label coordinates, respectively set the resource ratio dimension to the radar chart radius vector, the utilization efficiency ratio dimension to the height of the bar chart coordinate axis, and the job coverage ratio dimension to the heat map color scale index, assign three types of mapping parameters to each set of coordinate points, and establish a graphic coordinate mapping parameter table.

6. The method for visually assessing rehabilitation services according to claim 1, characterized in that, The specific steps for obtaining the visualization assessment chart set of rehabilitation services are as follows: S511: Based on the organization number field in the graphic coordinate mapping parameter table as the horizontal axis identifier, the resource ratio field is sequentially bound to the radar chart layer, and the radial region of the radar chart is constructed with the normalized value. For each organization node, vertices are generated in the polar coordinate system and closed regions are drawn according to the weights. After completing the graphic rendering, a resource radar rendering layer is generated. S512: Based on the sequence of organization nodes in the resource radar rendering layer, synchronously extract the utilization efficiency ratio field from the graphic coordinate mapping parameter table, map the normalized value to the Y-axis coordinate position of the bar chart layer, place the corresponding organization bar height according to the proportion, load the job coverage ratio field, map it to the color level value of the heat map layer, and assign it to the organization coordinate cell corresponding to the map layer to generate a multi-layer visual component set. S513: For each layer component in the multi-layer visual component set, set a time dimension scroll bar component, configure a layer switching button, aggregate all component structures into the same interactive interface visual area and establish linkage configuration rules to generate a rehabilitation service visualization assessment chart group.

7. A visualization assessment system for rehabilitation services, characterized in that, The system is used to implement the rehabilitation service visualization assessment method according to any one of claims 1-6, including: The indicator system design module is used to execute S1: obtain the dimension indicator entries in the TRIC and FRAME structures and collect data sources, identify the indicator codes, measurement units, and level classifications in the data sources, and set the collection methods and evaluation thresholds in combination with the indicator dimensions to generate a set of rehabilitation dimension indicator numbers; The unified data structure module is used to execute S2: based on the rehabilitation dimension indicator number set, obtain the structured questionnaire platform data and reporting database of rehabilitation service institutions in each region, detect the coding method, time format and unit type in each data, perform label comparison and unit conversion operations through field similarity, and generate a unified field sequence table; The indicator value integration module is used to execute S3: Based on the service usage frequency, resource allocation quantity, job function distribution, fiscal expenditure details and institution level identifier in the unified field sequence table, obtain the value distribution range under the same time dimension, perform weighted proportional allocation on resource allocation quantity and job function distribution, perform standardized interval normalization on service usage frequency and fiscal expenditure details, and synchronously group all normalization results according to the rehabilitation service institution level identifier to generate a grouped weighted normalization matrix; The graphic coordinate generation module is used to execute S4: based on the grouped weighted normalization matrix, it calculates the resource ratio, utilization efficiency ratio and job coverage ratio of each group, extracts the city code and connects it to the administrative region coordinate dictionary, sets the starting point of planar rendering according to the distribution order of rehabilitation service institutions, performs parameter mapping, generates a graphic coordinate mapping parameter table, sets the time dimension scrolling component and comparison item selection control, loads all graphics into the main view area of ​​the interactive interface, and generates a group of visual evaluation charts for rehabilitation services.

Images