Remote big data screening method based on standardized health record data set

Abstract

This invention discloses a remote big data screening method based on standardized health record datasets, relating to the field of electronic health record management technology. The method includes steps S1: data collection and standardization transformation; S2: centralized data processing and analysis; S3: analytical model construction and application; S4: presentation of screening results; and S5: data authenticity traceability mechanism. This remote big data screening method, based on standardized health record datasets, relies on existing resident electronic health record systems and public health information platforms in various regions. Technical personnel transform the raw data from the business database into a standardized dataset according to specifications. The standardized data is then automatically centrally screened and evaluated. This transforms the traditional manual, annually conducted evaluation work into a periodically and rapidly executable analysis model, significantly shortening the cycle from data readiness to evaluation results and solving the problem of data lag.

Description

Technical Field

[0001] This invention relates to the field of electronic health record management technology, specifically a remote big data screening method based on standardized health record datasets. Background Technology

[0002] Basic public health services are free public health interventions provided by the government to all residents, covering services such as immunization, chronic disease management, and the establishment of health records. To ensure service quality and effective resource allocation, service evaluation is usually used for quantitative assessment. Its significance lies in timely identification of service weaknesses, improvement of health management effectiveness, and providing data support for policy optimization and fund allocation, ultimately achieving the public health goal of reducing the disease burden and improving the health level of residents.

[0003] In the performance evaluation of basic public health services, traditional methods mainly rely on the manual collection and summarization of paper or electronic reports, such as Excel spreadsheets, from medical institutions at all levels, combined with on-site spot checks for assessment. However, this traditional method has several shortcomings: First, the data is severely outdated, and the annual evaluation cannot reflect real-time issues, leading to delayed management decisions. Second, the authenticity of the data is questionable. The reporting process at each level is prone to human manipulation and errors, making it difficult to verify. Third, the evaluation dimensions are too simplistic and crude, focusing more on "quantity" (such as the number of follow-ups) rather than "quality" and "effectiveness" (such as the rate of blood pressure control). This makes it impossible to scientifically measure the real health output of the service. This extensive and passive evaluation model is inefficient and makes it difficult to accurately identify and guide the weak links in the service. Summary of the Invention

[0004] The purpose of this invention is to provide a remote big data screening method based on standardized health record datasets to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a remote big data screening method based on standardized health record datasets, comprising the following steps: S1. Data Collection and Standardization Transformation: Based on existing resident electronic health record systems and basic public health service information platforms in various regions, we will collect electronic health record data such as personal basic information, physical examination records, and follow-up services of elderly people aged 65 and above, patients with hypertension, and patients with type 2 diabetes from medical institutions at all levels, and convert them into a standard dataset format to ensure data consistency and comparability. S2. Centralized Data Processing and Analysis: Centralized screening and evaluation analysis of standardized data includes checks on five aspects: data standardization, service quantity, record completeness, data correlation, and service standardization. This enables rapid data processing and analysis, improving the timeliness of evaluation. S3. Analysis Model Construction and Application: We will construct an intelligent evaluation algorithm system covering five core dimensions: standardization rate, deviation rate, correlation rate, completeness rate, and standardization rate. Through multi-angle and quantitative cross-analysis of service process data and result indicators, we will achieve comprehensive measurement of service quality. S4. Screening Results Presentation: Based on the screening results, an overall evaluation report is generated, including evaluation results in five dimensions: standard rate, deviation rate, correlation rate, completeness rate, and standardization rate. Abnormal data is automatically identified, and a detailed list of problems is generated so that users can quickly locate and resolve issues. S5. Data authenticity traceability mechanism: By establishing a mechanism that links upload dates with system operation logs, abnormal modification behaviors at time points after data reporting can be identified, providing a traceable technical basis for data authenticity verification.

[0006] Furthermore, in step S1, the specific operations of the standardization transformation are as follows: the data is cleaned, transformed, and formatted according to the standard dataset format requirements, and the transformed data is uploaded to the standard dataset. Data cleaning: Preprocessing the collected raw electronic health record data to remove or correct errors, duplicates or inconsistencies in the data; Data transformation: According to the requirements of standard dataset format, perform data type transformation (such as converting text data into numerical data) and data encoding transformation (such as encoding categorical variables into numerical codes) on the cleaned data; Data formatting: The converted data is formatted according to the standard dataset format, including setting the correct field names, field types, field lengths, etc.

[0007] Furthermore, in step S2, the screening content is as follows: Data standardization: Verify whether the electronic health record data format complies with national data standards in accordance with the "Technical Solution for Provincial System Interconnection of the National Primary Healthcare Comprehensive Management Platform" (v1.3.2); Service Quantity: Verify the number of electronic health records for key populations (elderly people aged 65 and above, patients with hypertension, and patients with type 2 diabetes) and compare them with the annual monitoring data; Record completeness: Verify whether the required fields for completeness in personal basic information, health check-up records, and follow-up records have been filled in; Data correlation: Based on the business logic consistency rules, the validity of personal basic information, health check-up records and follow-up records is correlated and queried through the individual's unique primary key; Service standardization: Verify whether the required fields in personal basic information, health check-up records, and follow-up records comply with preset rules.

[0008] Furthermore, in step S3, the standard rate is the proportion of data records that meet the standard to the total number of evaluation data records. The evaluation process for the standard rate is as follows: after the data is uploaded to the standard database, the system checks the data's value range and mandatory fields for compliance in accordance with the relevant provisions in the "Technical Solution for Provincial System Interconnection of the National Primary Healthcare Comprehensive Management Platform" to ensure the integrity and standardization of the data. The calculation formula is: Standard rate = (Number of data records that meet the standard / Total number of evaluation data records) × 100%.

[0009] Furthermore, in step S3, the deviation rate is the ratio of the deviation between the number of reports and the number of reports. The evaluation process for the deviation rate is as follows: after the data is uploaded to the standard database, the system compares the number of personal basic information in the uploaded data with the statistical data in the national basic public health service reports. The calculation formula is: deviation rate = ((number of reports - number of reports) / number of reports) × 100%.

[0010] Furthermore, in step S3, the correlation rate is the proportion of unique individuals with valid correlation records to the total number of reported individuals. The evaluation process for the correlation rate is as follows: After the data is uploaded to the standard database, the system performs a correlation query on the validity of personal basic information, health check-up records, and follow-up records based on the business logic consistency rules and through the individual's unique primary key, in order to verify the logical correlation of data between different business modules. The calculation formula is: Correlation rate = (unique individuals with valid correlation records / total number of reported individuals) × 100%.

[0011] Furthermore, in step S3, the completeness rate is the ratio of the total number of complete entries that have been filled out to the total number of complete entries that should be filled out. The evaluation process for the completeness rate is as follows: after the data is uploaded to the standard database, the system checks the required completeness fields in the personal basic information, health check-up records and follow-up records according to the preset completeness rules. The calculation formula is: completeness rate = (total number of complete entries that have been filled out / total number of complete entries that should be filled out) × 100%.

[0012] Furthermore, in step S3, the standardization rate is the proportion of the total number of items that meet the standards to the total number of reported items. The evaluation process for the standardization rate is as follows: after the data is uploaded to the standard database, the system checks the required standardization fields in personal basic information, health check-up records and follow-up records according to the preset integrity rules. The calculation formula is: standardization rate = (total number of items that meet the standards / total number of reported items) × 100%.

[0013] Furthermore, in step S4, the problem details list includes detailed information on abnormal data, including but not limited to data identifier, abnormal type, abnormal description, occurrence time, and related records, as follows: Data identifier: A unique identifier for anomalous data, such as a record ID or patient ID; Anomaly Type: Clearly indicate the type of anomaly, namely: data standardization anomaly (electronic health record data format does not conform to national data standards), service quantity anomaly (the number of electronic health records for key populations does not match the annual monitoring data), record completeness anomaly (mandatory fields for completeness in personal basic information, health examination records, and follow-up records are not filled in), data correlation anomaly (the logical connection between personal basic information, health examination records, and follow-up records is invalid), and service standardization anomaly (mandatory fields for standardization in personal basic information, health examination records, and follow-up records do not conform to preset rules). Anomaly Description: Describe the specific manifestation of the anomaly, such as "the data format does not conform to the XX requirements in the national data standard" or "the follow-up record is missing necessary XX information"; Time of occurrence: The time when the abnormal data was discovered or recorded; Related records: Other records related to the abnormal data, such as the patient's basic information and physical examination records, to facilitate tracing and analysis.

[0014] Furthermore, the specific operation of step S5 is as follows: the system records the data upload date and operation log, and uses the data upload date field as the key index. After completing the data quality screening, if data abnormalities or distortions are found, the system will trace and analyze the data operation records within a specific time period based on the upload date.

[0015] This invention provides a remote big data screening method based on standardized health record datasets, which has the following beneficial effects: 1. This invention relies on existing resident electronic health record systems and public health information platforms in various regions. Technical personnel convert raw data from the business database into a standardized dataset according to specifications, and then automatically perform centralized screening and evaluation analysis on the standardized data. This method separates the data conversion process from the system analysis process. After the data completes the standardization conversion and enters the standard set, the system initiates the processing flow. This transforms the traditional evaluation work, which relies on manual labor and is carried out annually, into a periodic and rapid analysis mode, significantly shortening the cycle from data readiness to the output of evaluation results and solving the problem of data lag.

[0016] 2. This invention uses the data upload date field as a key index. After data quality screening, if data anomalies or distortions are found, the data operation records within a specific time period can be traced and analyzed based on the upload date to help determine whether there have been any abnormal modifications to the data. By establishing a correlation mechanism between the upload date and the system operation log, this method can identify abnormal modification behaviors at abnormal time points after data reporting, providing a traceable technical basis for data authenticity verification.

[0017] 3. This invention constructs an intelligent evaluation algorithm system covering five core dimensions: standardization rate, deviation rate, correlation rate, completeness rate, and standardization rate. Through multi-faceted and quantitative cross-analysis of service process data and outcome indicators, it achieves a comprehensive measurement of service quality. This method, through the combination of algorithms across five dimensions, breaks through the traditional evaluation model that only focuses on "quantity" and "frequency," shifting towards a comprehensive assessment of service quality, standardization, and effectiveness. This allows for a more scientific measurement of the true output of services and solves the problem of a single evaluation dimension. Attached Figure Description

[0018] Figure 1 This is a schematic diagram illustrating the steps of the remote big data screening method based on a standardized health record dataset according to the present invention. Figure 2 This is a flowchart of the remote big data screening method for elderly people based on standardized health record datasets, as described in this invention; Figure 3 This is a flowchart of the standardized detection of the elderly using the remote big data screening method based on standardized health record datasets according to the present invention; Figure 4 This is a flowchart of the hypertension integrity detection method based on a standardized health record dataset for remote big data screening according to the present invention; Figure 5 This is a flowchart of the standardized hypertension detection method based on a standardized health record dataset for remote big data screening according to the present invention. Figure 6 This is a flowchart of the Phase 2 diabetes integrity detection method based on a standardized health record dataset, according to the present invention. Figure 7 This is a flowchart of the standardized detection of stage 2 diabetes using the remote big data screening method based on standardized health record datasets, as described in this invention. Detailed Implementation

[0019] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and should not be construed as limiting the scope of the invention.

[0020] Example 1 like Figure 1As shown, the remote big data screening method based on standardized health record datasets includes the following steps: S1. Data Collection and Standardization Transformation: Based on existing resident electronic health record systems and basic public health service information platforms in various regions, we collect electronic health record data, including personal basic information, physical examination records, and follow-up services, from medical institutions at all levels for elderly people aged 65 and above, patients with hypertension, and patients with type 2 diabetes. We then convert this data into a standard dataset format, which involves cleaning, converting, and formatting the data according to the requirements of the standard dataset format. The converted data is then uploaded to the standard dataset to ensure data consistency and comparability.

[0021] Data cleaning: Preprocessing the collected raw electronic health record data to remove or correct errors, duplicates, or inconsistencies in the data.

[0022] Data transformation: According to the requirements of the standard dataset format, perform data type transformation (such as converting text data into numerical data) and data encoding transformation (such as encoding categorical variables into numerical codes) on the cleaned data.

[0023] Data formatting: The converted data is formatted according to the standard dataset format, including setting the correct field names, field types, field lengths, etc.

[0024] In this embodiment, technicians transform the raw data in the business database according to the standard of the "Technical Solution for Provincial System Interconnection of National Primary Healthcare Comprehensive Management Platform" (v1.3.2) to form a standard dataset, ensuring that the data format, value range and required fields comply with national data standards.

[0025] S2. Centralized Data Processing and Analysis: Standardized data is centrally screened and evaluated, including checks on five aspects: data standardization, service quantity, record completeness, data correlation, and service standardization. This enables rapid data processing and analysis, improving the timeliness of evaluation. The specific screening content is as follows: Data standardization: Verify whether the electronic health record data format complies with national data standards in accordance with the "Technical Solution for Provincial System Interconnection of National Primary Healthcare Comprehensive Management Platform" (v1.3.2).

[0026] Service quantity: Verify the number of electronic health records for key populations (elderly people aged 65 and above, patients with hypertension, and patients with type 2 diabetes) and compare them with the annual monitoring data.

[0027] Record completeness: Verify whether the required fields for completeness in personal basic information, health check-up records, and follow-up records have been filled in.

[0028] Data correlation: Based on the business logic consistency rules, the validity of personal basic information, health check-up records and follow-up records is correlated and queried through the individual's unique primary key.

[0029] Service standardization: Verify whether the required fields in personal basic information, health check-up records, and follow-up records comply with preset rules.

[0030] S3. Analysis Model Construction and Application: We will construct an intelligent evaluation algorithm system covering five core dimensions: standardization rate, deviation rate, correlation rate, completeness rate, and standardization rate. Through multi-angle and quantitative cross-analysis of service process data and result indicators, we will achieve comprehensive measurement of service quality.

[0031] Standardization Rate: The proportion of data records that meet the standards out of the total number of evaluation data records. Evaluation Process for Standardization Rate: After data is uploaded to the standard database, the system, in accordance with the relevant provisions of the "Technical Solution for Provincial System Interconnection of the National Primary Healthcare Comprehensive Management Platform," checks the data's value range and mandatory fields to ensure data integrity and standardization. The calculation formula is: Standardization Rate = (Number of data records meeting the standards / Total number of evaluation data records) × 100%.

[0032] Deviation rate: The percentage difference between the number of reported data and the number of reported data. The evaluation process for the deviation rate is as follows: After the data is uploaded to the standard database, the system compares the amount of personal basic information in the uploaded data with the statistical data in the National Basic Public Health Service Reports. The calculation formula is: Deviation rate = ((Number of reported data - Number of reported data) / Number of reported data) × 100%.

[0033] Association Rate: The percentage of unique individuals with valid association records out of the total number of reported individuals. Association Rate Evaluation Process: After data is uploaded to the standard database, the system, based on business logic consistency rules, uses the individual's unique primary key to perform association queries on the validity of basic personal information, health check-up records, and follow-up records to verify the logical association between data from different business modules. The calculation formula is: Association Rate = (Number of unique individuals with valid association records / Total number of reported individuals) × 100%.

[0034] Completeness Rate: The percentage of complete fields already filled out out of the total number of complete fields that should be filled out. Completeness Rate Evaluation Process: After data is uploaded to the standard database, the system checks the required completeness fields in personal basic information, health check-up records, and follow-up records according to preset completeness rules. The calculation formula is: Completeness Rate = (Total number of complete fields already filled out / Total number of complete fields that should be filled out) × 100%.

[0035] Compliance Rate: The proportion of the total number of items that meet the standards to the total number of reported items. Evaluation Process of Compliance Rate: After the data is uploaded to the standard database, the system checks the required standardized fields in personal basic information, health examination records, and follow-up records according to preset integrity rules. The calculation formula is: Compliance Rate = (Total number of items that meet the standards / Total number of reported items) × 100%.

[0036] S4. Screening Results Presentation: Based on the screening results, a comprehensive evaluation report is generated, including evaluation results across five dimensions: standardization rate, deviation rate, correlation rate, completeness rate, and standardization rate. Abnormal data is automatically identified, and a detailed problem list is generated to help users quickly locate and resolve issues. The problem list contains detailed information about the abnormal data, including but not limited to data identifier, anomaly type, anomaly description, occurrence time, and related records, as detailed below: Data identifier: A unique identifier for anomalous data, such as a record ID or patient ID.

[0037] Anomaly Type: Clearly indicate the type of anomaly, namely: data standardization anomaly (electronic health record data format does not conform to national data standards), service quantity anomaly (the number of electronic health records for key populations does not match the annual monitoring data), record completeness anomaly (mandatory fields for completeness in personal basic information, health examination records, and follow-up records are not filled in), data correlation anomaly (the logical connection between personal basic information, health examination records, and follow-up records is invalid), and service standardization anomaly (mandatory fields for standardization in personal basic information, health examination records, and follow-up records do not conform to preset rules).

[0038] Anomaly Description: Describe the specific manifestation of the anomaly, such as "the data format does not conform to the XX requirements in the national data standard" or "the follow-up record is missing necessary XX information".

[0039] Time of occurrence: The time when the abnormal data was discovered or recorded.

[0040] Related records: Other records related to the abnormal data, such as the patient's basic information and physical examination records, to facilitate tracing and analysis.

[0041] S5. Data authenticity traceability mechanism: By establishing a correlation mechanism between upload date and system operation logs, abnormal modification behaviors at time points after data reporting are identified, providing traceable technical evidence for data authenticity verification. In this embodiment, the system records the data upload date and operation logs, and uses the data upload date field as a key index. After completing data quality screening, if data anomalies or distortions are found, the system performs traceability analysis on the data operation records within a specific time period based on the upload date.

[0042] Example 2 like Figure 2 As shown, this allows you to view the completeness and total data volume of elderly people aged 65 and above under the standardized management of districts and counties through big data screening and evaluation. Select the evaluation year, enter the urban and rural residents' health record ID, and click "Update Completeness System Detection Results." The background process performs a completeness check based on the built-in dictionary and rules.

[0043] Example 3 like Figure 3 As shown, select the evaluation year, enter the urban and rural residents' health record ID, and click "Update Completeness System Detection Results." The background will then perform a standardization check based on the built-in dictionary and rules.

[0044] Example 4 like Figure 4 As shown, select the evaluation year, enter the urban and rural residents' health record ID, and click "Update Completeness System Detection Results." The background will then perform a completeness check based on the built-in dictionary and rules.

[0045] Example 5 like Figure 5 As shown, select the evaluation year, enter the urban and rural residents' health record ID, and click "Update Completeness System Detection Results." The background will then perform a standardization check based on the built-in dictionary and rules.

[0046] Example 6 like Figure 6 As shown, select the evaluation year, enter the urban and rural residents' health record ID, and click "Update Completeness System Detection Results." The background will then perform a completeness check based on the built-in dictionary and rules.

[0047] Example 7 like Figure 7 As shown, select the evaluation year, enter the urban and rural residents' health record ID, and click "Update Completeness System Detection Results." The background will then perform a standardization check based on the built-in dictionary and rules.

[0048] The embodiments of the present invention are given for illustrative and descriptive purposes only, and are not intended to be exhaustive or to limit the invention to the forms disclosed. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to better illustrate the principles and practical application of the invention, and to enable those skilled in the art to understand the invention and to design various embodiments with various modifications suitable for a particular purpose.

Claims

1. A remote big data screening method based on standardized health record datasets, characterized in that, Includes the following steps: S1. Data Collection and Standardization Transformation: Based on the existing electronic health record systems and basic public health service information platforms in various regions, collect electronic health record data of elderly people aged 65 and above, patients with hypertension, and patients with type 2 diabetes from medical institutions at all levels, and transform them into a standard dataset format. S2. Centralized Data Processing and Analysis: Centralized screening and evaluation analysis of standardized data, including verification of five aspects: data standardization, service quantity, record completeness, data correlation, and service standardization. S3. Analysis Model Construction and Application: Construct an intelligent evaluation algorithm system covering five core dimensions: standard rate, deviation rate, correlation rate, completeness rate, and standardization rate. Through multi-angle and quantitative cross-analysis of service process data and result indicators, a comprehensive measurement of service quality can be achieved. S4. Screening Result Presentation: Generate an overall evaluation report based on the screening results, including evaluation results in five dimensions: standard rate, deviation rate, correlation rate, completeness rate, and standardization rate, and automatically identify abnormal data and generate a detailed list of issues; S5. Data authenticity traceability mechanism: By establishing a mechanism that links the upload date with the system operation log, abnormal modification behavior at abnormal time points after data reporting can be identified.

2. The remote big data screening method based on standardized health record datasets according to claim 1, characterized in that, In step S1, the specific operations of standardization transformation are as follows: the data is cleaned, transformed, and formatted according to the standard dataset format requirements, and the transformed data is uploaded to the standard dataset.

3. The remote big data screening method based on standardized health record datasets according to claim 1, characterized in that, In step S2, the screening content is as follows: Data standardization: Verify whether the electronic health record data format complies with national data standards; Service Quantity: Verify the number of electronic health records for key populations and compare them with annual monitoring data; Record completeness: Verify whether the required fields for completeness in personal basic information, health check-up records, and follow-up records have been filled in; Data correlation: Based on the business logic consistency rules, the validity of personal basic information, health check-up records and follow-up records is correlated and queried through the individual's unique primary key; Service standardization: Verify whether the required fields in personal basic information, health check-up records, and follow-up records comply with preset rules.

4. The remote big data screening method based on standardized health record datasets according to claim 1, characterized in that, In step S3, the standard rate is the proportion of data records that meet the standard to the total number of evaluation data records. The evaluation process for the standard rate is as follows: after the data is uploaded to the standard database, the system checks the data's value range and required fields for compliance according to relevant regulations. The calculation formula is: standard rate = (number of data records that meet the standard / total number of evaluation data records) × 100%.

5. The remote big data screening method based on standardized health record datasets according to claim 1, characterized in that, In step S3, the deviation rate is the ratio of the deviation between the number of reports and the number of reports. The evaluation process of the deviation rate is as follows: after the data is uploaded to the standard database, the system compares the number of personal basic information in the uploaded data with the statistical data in the national basic public health service reports. The calculation formula is: deviation rate = ((number of reports - number of reports) / number of reports) × 100%.

6. The remote big data screening method based on standardized health record datasets according to claim 1, characterized in that, In step S3, the correlation rate is the proportion of unique individuals with valid correlation records to the total number of reported individuals. The evaluation process for the correlation rate is as follows: After the data is uploaded to the standard database, the system performs a correlation query on the validity of personal basic information, health check-up records, and follow-up records based on the business logic consistency rules and the unique primary key of each individual to verify the logical correlation between data in different business modules. The calculation formula is: Correlation rate = (unique individuals with valid correlation records / total number of reported individuals) × 100%.

7. The remote big data screening method based on standardized health record datasets according to claim 1, characterized in that, In step S3, the completeness rate is the ratio of the total number of complete entries that have been filled out to the total number of complete entries that should be filled out. The evaluation process for the completeness rate is as follows: After the data is uploaded to the standard database, the system checks the required completeness fields in the personal basic information, health check-up records and follow-up records according to the preset completeness rules. The calculation formula is: Completeness rate = (Total number of complete entries that have been filled out / Total number of complete entries that should be filled out) × 100%.

8. The remote big data screening method based on a standardized health record dataset according to claim 1, characterized in that, In step S3, the standardization rate is the proportion of the total number of items that meet the standards to the total number of items reported. The evaluation process for the standardization rate is as follows: after the data is uploaded to the standard database, the system checks the required standardization fields in personal basic information, health check-up records and follow-up records according to the preset integrity rules. The calculation formula is: standardization rate = (total number of items that meet the standards / total number of items reported) × 100%.

9. The remote big data screening method based on standardized health record datasets according to claim 1, characterized in that, In step S4, the problem details list includes detailed information on abnormal data, including but not limited to data identifier, abnormal type, abnormal description, occurrence time, and related records, as follows: Data identifier: A unique identifier for anomalous data; Anomaly Type: Clearly indicate the type of anomaly, namely, data standardization anomaly, service quantity anomaly, record integrity anomaly, data correlation anomaly, and service specification anomaly; Exception Description: Describe the specific manifestations of the exception in detail; Time of occurrence: The time when the abnormal data was discovered or recorded; Related records: Other records related to the abnormal data.

10. The remote big data screening method based on a standardized health record dataset according to claim 1, characterized in that, The specific operation of step S5 is as follows: The system records the data upload date and operation log, and uses the data upload date field as the key index. After completing the data quality screening, if data abnormalities or distortions are found, the system will trace and analyze the data operation records within a specific time period based on the upload date.

Images