Radiographic image quality control management method and system

Abstract

The invention discloses a radiographic image quality control management method. The method comprises the following steps: S1, creating a quality control task; s2, performing multi-dimensional sampling on the check list according to a set sampling rule; s3, quality control experts are screened according to roles, and quality control tasks are distributed to the selected experts according to sampling results; s4, a quality control expert performs quality control operation on the distributed image or report based on the quality control template, and automatically classifies quality control result grades according to a preset score deduction range; s5, performing multi-dimensional statistical analysis on the completed quality control task; according to the method, repeated sample extraction and task conflicts are effectively avoided, the utilization efficiency of quality control resources is remarkably improved, and transparent double-blind quality control of mass image data is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of medical management systems, and particularly to a method and system for radiological image quality control management. Background Art

[0002] With the increasingly prominent core role of medical imaging technology in disease diagnosis, treatment, and efficacy evaluation, as a key quality control link, the quality of images and reports in the radiology department directly affects the accuracy of clinical decisions. Traditional quality control management mostly adopts manual sampling methods, which have inherent defects such as low efficiency, significant subjective bias, and insufficient sample coverage, and it is difficult to meet the needs of modern medical institutions for implementing standardized and transparent quality control of massive image data.

[0003] Chinese invention patent CN114927199A (publication date: August 19, 2022) discloses a medical image quality control system based on cloud computing, which creates sampling tasks through an administrator workstation and assigns them to quality control experts for scoring. Although this solution realizes the digital management of quality control tasks, there are still the following technical bottlenecks: 1) The sampling rules are single, lacking a multi-dimensional mutually exclusive sampling mechanism for examination categories, equipment types, and anatomical parts, resulting in the omission or duplication of key quality control samples; 2) Task assignment depends on manual experience, and an intelligent assignment model based on expert load thresholds and dynamic ratio adjustment is not established, which is prone to uneven resource allocation; 3) The quality control evaluation dimensions are fixed, lacking a defect factor weight optimization mechanism based on historical data, and it is difficult to achieve the dynamic evolution of quality control standards.

[0004] Especially when dealing with scenarios of multi-hospital collaboration and complex examination type intersections, the existing system cannot effectively prevent sample conflicts between image quality control and report quality control tasks, resulting in duplicate quality control and resource waste. In addition, during the implementation of traditional double-blind quality control, there is a lack of a real-time tracking mechanism for the status of a large number of examination forms, making it difficult to ensure the integrity and traceability of sampling data. Summary of the Invention

[0005] The purpose of the present invention is to address the problems existing in the prior art, and provide a method and system for radiological image quality control management, which can effectively avoid duplicate sample extraction and task conflicts, significantly improve the utilization efficiency of quality control resources, and achieve transparent double-blind quality control of massive image data.

[0006] To achieve the above purpose, the technical solution adopted by the present invention is:

[0007] A method for radiological image quality control management, comprising the following steps:

[0008] S1. Create a quality control task;

[0009] S2. Perform multi-dimensional sampling on examination forms according to the set sampling rules;

[0010] S3. Screen quality control experts according to their roles, and assign quality control tasks to the selected experts based on the sampling results;

[0011] S4. Quality control experts perform quality control operations on the assigned images or reports based on the quality control template, and automatically classify the quality control result levels according to the preset deduction range;

[0012] S5. Conduct multi-dimensional statistical analysis on the completed quality control tasks.

[0013] Step S1 includes:

[0014] S11. Set the quality control task attributes, including task name, task category, start time, and end time;

[0015] S12. Dynamically display the task status information in the task list, including the total sampling volume, the number of quality control tasks completed, the task progress percentage, and the task status mark;

[0016] S13. Generate a task details view, including the distribution of the number of quality controlled / uncontrolled items, the list of quality control experts and their task completion rates, the trend chart of quality control score distribution, and the detailed checklist.

[0017] Step S2 includes:

[0018] S21. Configure the quality control sampling rules, including the hospital area scope, date range, examination category, examination technician, examination equipment, examination site grouping, radiology professional title level, and random sampling mode;

[0019] S22. After applying the sampling rules, display the number of samples filtered by the rules in real time, and generate the final sampling results based on the following mutually exclusive principles:

[0020] The inspection list that has been sampled for the image quality control task will no longer participate in the sampling of subsequent image quality control tasks, but can be sampled for the report quality control task;

[0021] The inspection list that has been sampled for the report quality control task will no longer participate in the sampling of subsequent report quality control tasks, but can be sampled for the image quality control task.

[0022] Step S3 includes:

[0023] S31. Screen quality control experts according to the preset role permissions, support single-selection or multi-selection operations, and display the number of selected experts in real time;

[0024] S32. Set the workload upper limit threshold for each quality control expert, where the workload upper limit includes the maximum number of assignable inspection lists or the maximum task load duration;

[0025] S33. Automatically calculate the amount of quality control tasks that should be assigned to each expert according to the sampling results. When the remaining assignable amount of an expert is lower than the amount of tasks to be assigned, dynamically assign them proportionally;

[0026] S34. Visualize and display on the task assignment interface the comparison relationship among the task volume already undertaken by experts, the remaining allocable volume, and the number of tasks assigned this time.

[0027] S35. For imaging quality control tasks, the assigned inspection sheets are automatically marked as the "sampled" status, and are prohibited from being re-assigned to other imaging quality control experts; for report quality control tasks, cross-task type repeated assignment is allowed, but repeated processing by the same expert is restricted.

[0028] Step S4 includes:

[0029] S41. The quality control expert matches the quality control template according to the task type, and the template includes a list of quality control factors associated with the inspection category and the corresponding deduction rules.

[0030] S42. Retrieve the imaging or report content of the sampled inspection sheet, select the factor items that trigger quality control problems in the template, perform item-by-item deduction and generate comments.

[0031] S43. Real-time summarize the total deduction score, and automatically classify the quality level according to the deduction interval threshold preset by the scoring classification module.

[0032] S44. The quality control result includes raw data, deduction details, manual evaluation, and system-determined level, and is stored in association with the inspection unit data.

[0033] Step S5 includes:

[0034] S51. For imaging quality control tasks, perform cross-statistical analysis according to at least the following three dimensions:

[0035] Based on the dimension of the inspection technician, including statistics on the distribution of quality control deduction items, the proportion of excellent, good, and poor evaluations, and the high-frequency problem factors of different technicians;

[0036] Based on the dimension of the inspection category, including comparing the differences in the quality control pass rates of different imaging examination types such as CT, MR, CR, and RF;

[0037] Based on the dimension of the quality control factor, including analyzing the top 5 quality control defect factors with the highest occurrence frequency in each inspection site group;

[0038] S52. For report quality control tasks, perform correlation analysis according to at least the following three dimensions:

[0039] Based on the dimension of the reporting doctor, including statistics on the distribution of report standardization scores and the clustering of error types of doctors at different professional title levels;

[0040] Based on the dimension of the inspection site, including analyzing the defect rates of report descriptions in different anatomical sites such as the nervous system and the skeletal system;

[0041] Trend analysis based on time dimension, including tracking the dynamic change trends of key quality control indicators at weekly / monthly granularity;

[0042] S53. Generate visual statistical reports, including multi-dimensional data pivot tables, quality control KPI compliance matrix, and defect heat maps;

[0043] Among them, the multi-dimensional pivot table supports multi-level drill-down analysis by hospital area, equipment model, and radiology professional title;

[0044] The quality control KPI achievement matrix shows the achievement of image quality pass rate and report integrity indicators for each inspection category;

[0045] Defect heatmaps show the distribution density of quality control problems in different inspection rooms / equipment through spatial mapping;

[0046] S54. Establish a quality control traceability mechanism to automatically associate the original inspection sheets, quality control evaluation records and corresponding quality control expert feedback with the systemic quality problems discovered through statistics.

[0047] Step S6: Constructing a dynamic feedback optimization mechanism, specifically including:

[0048] S61. Based on the statistical analysis results of step S5, identify high-frequency quality control defect factors and associated dimension combinations, and generate a risk weight matrix;

[0049] S62. When configuring the subsequent sampling rules, an intelligent sampling optimization engine is introduced to automatically adjust the sampling ratio of each dimension according to the risk weight matrix;

[0050] S63. For the inspection technicians or reporting doctors whose quality control pass rate is lower than the threshold in three consecutive statistical analyses, the generation of special quality control tasks is automatically triggered, and the sampling ratio of the relevant inspection sheets of this person in the sampling rules is forcibly set to 100%.

[0051] A radiological image quality control management system, comprising:

[0052] Task creation module, used to create quality control tasks;

[0053] The sampling module is used to perform multi-dimensional sampling on the checklist according to the set sampling rules;

[0054] The task assignment module is used to screen quality control experts according to their roles and assign quality control tasks to selected experts based on sampling results;

[0055] The quality control operation module is used by quality control experts to perform quality control operations on assigned images or reports based on quality control templates, and automatically classify the quality control results according to the preset deduction range;

[0056] The statistical analysis module is used to perform multi-dimensional statistical analysis on completed quality control tasks.

[0057] A computer-readable storage medium stores a computer program therein. When the computer program is executed by a processor, the above method steps are implemented.

[0058] An electronic device includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the above method steps are implemented.

[0059] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0060] A full-process quality control management system based on multi-dimensional mutually exclusive sampling rules and dynamic task allocation is constructed. Precise sampling is achieved by setting multiple filtering conditions such as hospital areas, inspection categories, and equipment. Combining the mutual exclusion principle and double anti-duplication mechanism of imaging / report quality control tasks, sample duplicate extraction and task conflicts are effectively avoided.

[0061] An intelligent expert task allocation algorithm is adopted to dynamically adjust the allocation ratio according to the workload threshold, and a visual task tracking interface is constructed, significantly improving the utilization efficiency of quality control resources and realizing transparent double-blind quality control of massive imaging data.

[0062] Through the quality control factor library and automatic deduction rules associated with inspection categories, subjective experience is transformed into a standardized evaluation process. The deduction interval threshold of the scoring classification module is introduced in the quality control process to realize the automatic determination of quality grades, eliminate human scoring biases, and construct a traceable quality control evidence chain through the strong association storage mechanism of imaging data and quality control results, making the quality control evaluation process verifiable and repeatable. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0064] Figure 1 It is a diagram of the quality control task setting interface in the embodiment of the present application;

[0065] Figure 2 It is a diagram of the task list interface in the embodiment of the present application;

[0066] Figure 3 It is a diagram of the task list interface in the embodiment of the present application;

[0067] Figure 4It is the interface diagram of the checklist details list in the embodiment of this application;

[0068] Figure 5 It is the interface diagram of the quality control sampling rule setting in the embodiment of this application;

[0069] Figure 6 It is the interface diagram of the quality control sampling rule setting in the embodiment of this application;

[0070] Figure 7 It is the interface diagram of the quality control expert setting in the embodiment of this application;

[0071] Figure 8 It is the interface diagram of the quality control template list in the embodiment of this application;

[0072] Figure 9 It is the interface diagram of the quality control template setting in the embodiment of this application;

[0073] Figure 10 It is the interface diagram of the spot-check image quality control list in the embodiment of this application;

[0074] Figure 11 It is the interface diagram of the spot-check report quality control list in the embodiment of this application;

[0075] Figure 12 It is the interface diagram of the spot-check image quality control operation in the embodiment of this application. Detailed implementation manners

[0076] Next, the technical solutions of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0077] The magnitude of the sequence numbers of the steps in the description of this application does not mean the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of this application.

[0078] In the description of the specification and the appended claims of the present application, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and should not be construed as indicating or implying relative importance. It should also be understood that although the terms "first", "second", etc. are used in the text to describe various elements in some embodiments of the present application, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, the first table can be named the second table, and similarly, the second table can be named the first table, without departing from the scope of the various described embodiments. The first table and the second table are both tables, but they are not the same table.

[0079] The reference to "one embodiment" or "some embodiments" etc. described in the specification of the present application means that a specific feature, structure, or characteristic described in connection with that embodiment is included in one or more embodiments of the present application. Thus, the statements "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in another way. The terms "comprising", "including", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in another way.

[0080] Chinese invention patent CN114927199A (publication date: August 19, 2022) discloses a medical image quality control system based on cloud computing, which creates sampling tasks through an administrator workstation and assigns them to quality control experts for scoring. Although this solution realizes the digital management of quality control tasks, there are still the following technical bottlenecks: 1) The sampling rules are single, lacking a multi-dimensional mutually exclusive sampling mechanism for examination categories, equipment types, and anatomical parts, resulting in the omission or duplication of key quality control samples; 2) Task assignment depends on manual experience, and an intelligent assignment model based on expert load thresholds and dynamic ratio adjustment has not been established, which is likely to cause uneven resource allocation; 3) The quality control evaluation dimensions are fixed, lacking a defect factor weight optimization mechanism based on historical data, and it is difficult to achieve the dynamic evolution of quality control standards.

[0081] Especially when dealing with scenarios of multi-hospital collaboration and complex examination type intersections, the existing system cannot effectively prevent sample conflicts between image quality control and report quality control tasks, resulting in duplicate quality control and resource waste. In addition, during the implementation of traditional double-blind quality control, there is a lack of a real-time tracking mechanism for the status of a large number of examination sheets, making it difficult to ensure the integrity and traceability of sampling data.

[0082] To address the above technical problems, the embodiments of the present application provide a method for radiological image quality control management, including the following steps:

[0083] S1. Create a quality control task.

[0084] In some embodiments, step S1 includes the following steps S11 - S13.

[0085] S11. As Figure 1 shown, set the quality control task attributes, including task name, task category, start time, and end time.

[0086] S12. As Figure 2 , 3 shown, dynamically display the task status information in the task list, including the total sampling quantity, the number of completed quality controls, the task progress percentage, and the task status flag.

[0087] S13. As Figure 4 shown, generate a task details view, including the distribution of the quantity of quality - controlled / non - quality - controlled, the list of quality control experts and their task completion rates, the trend chart of quality control score distribution, and the detailed checklist.

[0088] Through the above method steps, create a quality control task, set the task attributes in detail, dynamically display the task status information, and generate a task details view, realizing the comprehensive management and refined monitoring of the quality control task, improving the quality control efficiency, and ensuring the transparency and traceability of the quality control work.

[0089] S2. Perform multi - dimensional sampling on the checklist according to the set sampling rules.

[0090] In some embodiments, step S2 includes the following steps S21 - S22.

[0091] S21. As Figure 5 , 6 shown, configure the quality control sampling rules, including the hospital area scope, date range, examination category, examination technician, examination equipment, examination site grouping, radiology professional title level, and random sampling mode.

[0092] S22. After applying the sampling rules, display the number of samples filtered by the rules in real - time, and generate the final sampling result based on the following mutually exclusive principles:

[0093] The checklist that has been sampled by the imaging quality control task will no longer participate in the sampling of subsequent imaging quality control tasks, but can be sampled by the report quality control task;

[0094] The checklist that has been sampled by the report quality control task will no longer participate in the sampling of subsequent report quality control tasks, but can be sampled by the imaging quality control task.

[0095] Through the above method steps, the sampling rules are flexibly configured and the number of filtered samples is displayed in real time. At the same time, the mutually exclusive sampling of the checklist in the image quality control and report quality control tasks is ensured, so as to achieve efficient and accurate sampling of the checklist, avoid repeated quality control, improve the utilization rate of quality control resources, and ensure the objectivity and accuracy of the quality control results.

[0096] S3. Screen quality control experts according to roles, and allocate quality control tasks to the selected experts according to the sampling results.

[0097] In some embodiments, step S3 includes the following steps S31 - S35.

[0098] S31. As Figure 7 shown, screen quality control experts according to the preset role permissions, support single - selection or multi - selection operations, and display the number of selected experts in real time;

[0099] S32. As Figure 7 shown, set an upper limit threshold for the workload of each quality control expert, and the workload upper limit includes the maximum number of checklists that can be assigned or the maximum task load duration;

[0100] S33. Automatically calculate the amount of quality control tasks that should be assigned to each expert according to the sampling results. When the remaining assignable amount of an expert is lower than the amount of tasks to be assigned, it is dynamically assigned proportionally;

[0101] S34. Visually display the comparison relationship between the amount of tasks already undertaken by the expert, the remaining assignable amount, and the number of tasks assigned this time in the task assignment interface;

[0102] S35. For the image quality control task, the assigned checklist is automatically marked as the "sampled" status and is prohibited from being re - assigned to other image quality control experts; for the report quality control task, cross - task - type repeated assignment is allowed but repeated processing by the same expert is restricted.

[0103] Through the above method steps, quality control experts are screened according to the preset role permissions, the workload upper limit is flexibly set, quality control tasks are automatically assigned according to the sampling results, the task assignment situation is visually displayed in the task assignment interface, and a dual anti - repetition mechanism is established, ensuring the reasonable distribution of quality control tasks, improving the quality control efficiency, avoiding repeated task assignment, and guaranteeing the independence and accuracy of the quality control results.

[0104] S4. Quality control experts perform quality control operations on the assigned images or reports based on the quality control template, and automatically classify the quality control result levels according to the preset deduction range.

[0105] In some embodiments, step S4 includes the following steps S41 - S44.

[0106] S41. As Figure 8 and Figure 9As shown in the figure, the quality control expert matches the quality control template according to the task type, and the template includes a list of quality control factors associated with the inspection category and the corresponding deduction rules;

[0107] S42. For example, Figures 10 to 12 As shown in the figure, retrieve the image or report content of the sampled inspection form, select the factor items that trigger quality control problems in the template, execute item-by-item deduction and generate comments;

[0108] S43. Real-time summarize the total deduction score, and automatically classify the quality level according to the deduction interval threshold preset by the scoring classification module;

[0109] S44. The quality control result includes the original data, deduction details, manual evaluation and system-determined level, and is associated with the inspection unit data for storage.

[0110] Through the above method steps, the standardization and efficiency of the quality control process are ensured; by matching the quality control template, the quality control expert can quickly locate the quality control factors and deduction rules for the inspection category, achieve accurate item-by-item deduction and generate comments. Real-time summarize the deductions and automatically classify the quality control result level, improving the quality control efficiency and ensuring the objectivity and consistency of the quality control results. At the same time, comprehensively record the quality control results and associate them with the inspection unit data for storage, providing strong support for subsequent quality control traceability, analysis and improvement, and further enhancing the reliability and effectiveness of radiological image quality control.

[0111] S5. Conduct multi-dimensional statistical analysis on the completed quality control tasks.

[0112] In some embodiments, step S5 includes the following steps S51-S54.

[0113] S51. For the image quality control task, conduct cross-statistical analysis according to at least the following three dimensions:

[0114] Based on the dimension of the inspection technician, including counting the distribution of quality control deduction items, the proportion of excellent, good and poor evaluations, and the high-frequency problem factors of different technicians;

[0115] Based on the dimension of the inspection category, including comparing the differences in the quality control passing rates of different imaging inspection types such as CT, MR, CR, and RF;

[0116] Based on the dimension of the quality control factor, including analyzing the top 5 quality control defect factors with the highest occurrence frequency in each inspection site group.

[0117] S52. For the report quality control task, conduct correlation analysis according to at least the following three dimensions:

[0118] Based on the dimension of the reporting doctor, including counting the distribution of report standardization scores and the clustering of error types of doctors at different professional title levels;

[0119] Based on the dimensions of the examination site, including analysis of the report description defect rate of different anatomical sites such as the nervous system and skeletal system;

[0120] Trend analysis based on the time dimension, including tracking the dynamic change trends of key quality control indicators at weekly / monthly granularity.

[0121] S53. Generate visual statistical reports, including multi-dimensional data pivot tables, quality control KPI compliance matrix, and defect heat maps;

[0122] Among them, the multidimensional pivot table supports multi-level drill-down analysis by hospital area, equipment model, and radiology professional title;

[0123] The quality control KPI compliance matrix shows the achievement of image quality qualification rate and report integrity indicators for each inspection category;

[0124] The defect heat map shows the distribution density of quality control problems in different inspection rooms / equipment through spatial mapping.

[0125] S54. Establish a quality control traceability mechanism to automatically associate the original inspection sheets, quality control evaluation records and corresponding quality control expert feedback with the systemic quality problems discovered through statistics.

[0126] The above methods and steps enable comprehensive and in-depth exploration of valuable information within quality control data. Through cross-sectional statistical and correlation analysis across multiple dimensions, quality control issues and their root causes can be precisely identified. Furthermore, the generated visual statistical reports provide an intuitive and clear presentation for quality control management, helping managers quickly understand quality control status and make decisions.

[0127] S6. Build a dynamic feedback optimization mechanism.

[0128] In some embodiments, step S6 specifically includes the following steps S61-S64.

[0129] S61. Based on the statistical analysis results of step S5, identify high-frequency quality control defect factors and associated dimension combinations, and generate a risk weight matrix.

[0130] S62. When configuring subsequent sampling rules, an intelligent sampling optimization engine is introduced to automatically adjust the sampling ratio of each dimension according to the risk weight matrix, so that the sampling probability of high-risk inspection categories or technicians is increased by 30%-50%.

[0131] S63. For an inspection technician or reporting doctor whose quality control pass rate is lower than the threshold in three consecutive statistical analyses, a special quality control task is automatically triggered to be generated, and the sampling ratio of the relevant inspection sheets of the person in the sampling rules is forcibly set to 100%.

[0132] S64. Establish a cross-task knowledge transfer channel, dynamically push the high-frequency deduction factors in the historical quality control template to the recommended list of the newly created template, and predict the potential correlation of the newly added quality control factors based on the Bayesian algorithm.

[0133] Specifically, in some embodiments, step S64 specifically includes the following steps:

[0134] S641. The system builds a historical quality control template database to store the quality control template data of all completed tasks, including:

[0135] The list of quality control factors associated with each inspection category (such as CT, MR, etc.);

[0136] The triggering frequency, deduction weight of each quality control factor in historical tasks, and the corresponding inspection site grouping;

[0137] The manual marking information of quality control experts for quality control factors (such as whether it is a key defect).

[0138] S642. Automatically screen high-frequency deduction factors according to preset thresholds (such as triggering frequency ≥ N times / month, deduction weight ≥ X points);

[0139] Cluster the factors by inspection category and inspection site grouping to generate a factor set {F1, F2,..., Fn}.

[0140] S643. When the user creates a new quality control template and adds a new quality control factor Fx, the system calls the Bayesian model, and the system automatically retrieves the template data of the same or similar inspection categories in the historical template database.

[0141] S644. Calculate the comprehensive priority score based on the factor triggering frequency, deduction weight, and manual marking information, and generate a recommended list in descending order of the score; the recommended list is dynamically embedded in the new template editing interface, supporting one-key import operation.

[0142] S645. Use the naive Bayesian classifier to calculate the conditional probability of the newly added factor Fx and the historical factor set {F1, F2,..., Fn}:

[0143]

[0144] S646. When the user adds a new quality control factor Fx in the new template, the system calls the Bayesian model to calculate the correlation probability {P1, P2,..., Pn} between Fx and each factor in the recommended list.

[0145] S647. If there exists Pi ≥ threshold θ (such as θ = 0.7), then highlight Fi in the recommended list and mark the predicted correlation relationship.

[0146] Through the above methods and steps, it is possible to intelligently identify high-frequency quality control defects and risk points based on statistical analysis of historical quality control data. By adjusting the sampling ratio and optimizing the sampling rules, it is possible to achieve focused monitoring of high-risk inspection categories or technicians. At the same time, for individuals with consistently low quality control pass rates, special quality control tasks are automatically triggered to ensure that problems are resolved in a timely and effective manner. In addition, the establishment of a cross-task knowledge transfer channel promotes the sharing and inheritance of quality control knowledge and experience, provides strong support for the continuous optimization and improvement of quality control templates, and thus promotes the continuous improvement and optimization of radiological imaging quality control work.

[0147] In summary, this application has constructed a full-process quality control management system based on multi-dimensional mutually exclusive sampling rules and dynamic task allocation. It achieves accurate sampling by setting multiple filtering conditions such as hospital area, inspection category, and equipment. Combined with the mutual exclusivity principle and double anti-duplication mechanism of image / report quality control tasks, it effectively avoids repeated sample extraction and task conflicts.

[0148] It adopts an intelligent expert task allocation algorithm, dynamically adjusts the allocation ratio according to the workload threshold, and builds a visual task tracking interface to significantly improve the efficiency of quality control resource utilization. Compared with the traditional manual mode, it reduces manpower consumption by more than 60%, and realizes transparent double-blind quality control of massive imaging data.

[0149] By checking the quality control factor library associated with categories and the automatic deduction rules, subjective experience is converted into a standardized evaluation process. The deduction interval threshold of the scoring classification module is introduced into the quality control process to realize automatic determination of quality levels and eliminate human scoring bias. Through the strong association storage mechanism between image data and quality control results, a traceable quality control evidence chain is constructed, making the quality control evaluation process verifiable and repeatable.

[0150] In a second aspect of an embodiment of the present application, a radiological image quality control management system is provided, comprising:

[0151] Task creation module, used to create quality control tasks;

[0152] The sampling module is used to perform multi-dimensional sampling on the checklist according to the set sampling rules;

[0153] The task assignment module is used to screen quality control experts according to their roles and assign quality control tasks to selected experts based on sampling results;

[0154] The quality control operation module is used by quality control experts to perform quality control operations on assigned images or reports based on quality control templates, and automatically classify the quality control results according to the preset deduction range;

[0155] The statistical analysis module is used to perform multi-dimensional statistical analysis on completed quality control tasks.

[0156] In some embodiments, the task creation module includes:

[0157] A task attribute setting unit for setting quality control task attributes, including task name, task category, start time, and end time;

[0158] A task status display unit for dynamically displaying task status information in the task list, including total sampling volume, number of completed quality control tasks, task progress percentage, and task status marker;

[0159] A task details generation unit for generating a task details view, including the distribution of the number of quality controlled / unquality controlled, the list of quality control experts and their task completion rates, the trend chart of quality control score distribution, and the detailed list of inspection sheets.

[0160] In some embodiments, the sampling module includes:

[0161] A sampling rule configuration unit for configuring quality control sampling rules, including hospital area scope, date range, inspection category, inspection technician, inspection equipment, inspection site grouping, radiology professional title level, and random sampling mode;

[0162] A sample quantity display unit for real-time displaying the number of samples filtered by the rule after applying the sampling rule;

[0163] A mutually exclusive principle application unit for generating the final sampling result based on the mutually exclusive principle, where:

[0164] The inspection sheets that have been sampled by the image quality control task will no longer participate in the sampling of subsequent image quality control tasks, but can be sampled by the report quality control task;

[0165] The inspection sheets that have been sampled by the report quality control task will no longer participate in the sampling of subsequent report quality control tasks, but can be sampled by the image quality control task.

[0166] In some embodiments, the task assignment module includes:

[0167] An expert screening unit for screening quality control experts according to preset role permissions, supporting single-selection or multi-selection operations, and real-time displaying the number of selected experts;

[0168] A workload setting unit for setting the workload upper limit threshold for each quality control expert, where the workload upper limit includes the maximum number of assignable inspection sheets or the maximum task load duration;

[0169] A task quantity calculation unit for automatically calculating the quality control task quantity to be assigned to each expert according to the sampling result. When the remaining assignable quantity of an expert is lower than the quantity of tasks to be assigned, it is dynamically assigned proportionally;

[0170] A task assignment interface unit for visually displaying the comparison relationship of the task volume already undertaken by an expert, the remaining assignable volume, and the number of tasks assigned this time on the task assignment interface;

[0171] A duplicate prevention mechanism unit for establishing a dual duplicate prevention mechanism, where:

[0172] For image quality control tasks, the assigned inspection list is automatically marked as the "sampled" status and is prohibited from being re-assigned to other image quality control experts;

[0173] For report quality control tasks, cross-task type duplicate assignment is allowed, but duplicate processing by the same expert is restricted.

[0174] In some embodiments, the quality control operation module includes:

[0175] A template matching unit for a quality control expert to match a quality control template according to the task type, where the template contains a list of quality control factors associated with the inspection category and the corresponding deduction rules;

[0176] A quality control execution unit for retrieving the image or report content of the sampled inspection list, selecting the factor items that trigger quality control problems in the template, performing item-by-item deduction and generating comments;

[0177] A deduction summary unit for real-time summarizing the total deduction value and automatically classifying the quality level according to the deduction interval threshold preset by the scoring classification module;

[0178] A result storage unit for storing the quality control results, where the quality control results include the original data, the deduction details, the manual evaluation, and the system-determined level, and are stored in association with the data of the inspection unit.

[0179] In some embodiments, the statistical analysis module includes:

[0180] An image quality control analysis unit for performing cross-statistical analysis on image quality control tasks in at least the following three dimensions:

[0181] Based on the dimension of the inspection technician, including counting the distribution of quality control deduction items, the proportion of excellent, good, and poor evaluations, and the high-frequency problem factors of different technicians;

[0182] Based on the dimension of the inspection category, including comparing the differences in the quality control pass rates of different image inspection types such as CT, MR, CR, and RF;

[0183] Based on the dimension of the quality control factor, including analyzing the top 5 quality control defect factors with the highest occurrence frequency in each inspection site group;

[0184] A report quality control analysis unit for performing correlation analysis on report quality control tasks in at least the following three dimensions:

[0185] Based on the dimension of reporting doctors, including statistically analyzing the distribution of reporting compliance scores and clustering of error types for doctors at different professional title levels;

[0186] Based on the dimension of examination sites, including analyzing the defect rates of report descriptions for different anatomical sites such as the nervous system and skeletal system;

[0187] Trend analysis based on the time dimension, including tracking the dynamic change trends of key quality control indicators at weekly / monthly granularity;

[0188] Visual report generation unit, used to generate visual statistical reports, including multi-dimensional data pivot tables, quality control KPI compliance matrices, and defect heat maps;

[0189] Among them, the multi-dimensional data pivot table supports multi-level drill-down analysis at the levels of hospital areas, equipment models, and radiology professional titles;

[0190] The quality control KPI compliance matrix shows the achievement degrees of each examination category in terms of image quality pass rate and report integrity indicators;

[0191] The defect heat map shows the distribution density of quality control problems in different examination rooms / equipment through spatial mapping;

[0192] Quality control traceability unit, used to establish a quality control traceability mechanism, and automatically associate the original examination forms, quality control evaluation records, and corresponding feedback from quality control experts with systematic quality problems found through statistics.

[0193] In the third aspect of the embodiments of the present application, a computer-readable storage medium is provided. A computer program is stored in the computer-readable storage medium. When the computer program is executed by a processor, the above method steps are implemented.

[0194] In the fourth aspect of the embodiments of the present application, an electronic device is provided, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the above method steps are implemented.

[0195] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for quality control management of radiological images, characterized in that, It includes the following steps: S1. Create a quality control task; S2. Perform multi-dimensional sampling on the inspection list according to the set sampling rules; S3. Screen quality control experts according to roles, and allocate quality control tasks to the selected experts according to the sampling results; S4. Quality control experts perform quality control operations on the allocated images or reports based on the quality control template, and automatically classify the quality control result levels according to the preset deduction range; S5. Conduct multi-dimensional statistical analysis on the completed quality control tasks.

2. The method for radiological imaging quality control management according to claim 1, wherein Step S1 includes: S11. Set the quality control task attributes, including task name, task category, start time, and end time; S12. Dynamically display the task status information in the task list, including the total sampling volume, the number of quality controls completed, the task progress percentage, and the task status mark; S13. Generate a task details view, including the distribution of the number of quality controlled / uncontrolled, the list of quality control experts and their task completion rates, the quality control score distribution trend chart, and the inspection list details.

3. A radiological image quality control management method according to claim 1, characterized in that, Step S2 includes: S21. Configure the quality control sampling rules, including hospital area range, date range, inspection category, inspection technician, inspection equipment, inspection site grouping, radiology title level, and random sampling mode; S22. After applying the sampling rules, display the number of samples filtered by the rules in real time, and generate the final sampling result based on the following mutually exclusive principles: The inspection list that has been sampled by the image quality control task will no longer participate in the sampling of subsequent image quality control tasks, but can be sampled by the report quality control task; The inspection list that has been sampled by the report quality control task will no longer participate in the sampling of subsequent report quality control tasks, but can be sampled by the image quality control task.

4. A method for radiological image quality control management according to claim 1, characterized in that, Step S3 includes: S31. Screen quality control experts according to the preset role permissions, support single or multiple selection operations, and display the number of selected experts in real time; S32. Set the workload upper limit threshold for each quality control expert, and the workload upper limit includes the maximum number of inspection lists that can be allocated or the maximum task load duration; S33. Automatically calculate the amount of quality control tasks that should be allocated to each expert according to the sampling results. When the remaining allocable amount of an expert is lower than the amount of tasks to be allocated, allocate dynamically according to the proportion; S34. Visually display the comparison relationship between the amount of tasks already undertaken by the expert, the remaining allocable amount, and the number of tasks allocated this time in the task assignment interface; S35. For the image quality control task, the allocated inspection list is automatically marked as the "sampled" status and is prohibited from being re-allocated to other image quality control experts; for the report quality control task, cross-task type repeated allocation is allowed, but repeated processing by the same expert is restricted.

5. The method for radiological image quality control management according to claim 1, wherein Step S4 includes: S41. Quality control experts match the quality control template according to the task type, and the template contains a list of quality control factors associated with the inspection category and the corresponding deduction rules; S42. Retrieve the image or report content of the sampled inspection list, select the factor items that trigger quality control problems in the template, perform item-by-item deductions and generate comments; S43. Summarize the total deduction value in real time, and automatically classify the quality level according to the deduction interval threshold preset by the scoring classification module; S44. The quality control result includes the original data, the deduction details, the manual evaluation, and the system-determined level, and is associated and stored with the inspection unit data.

6. The method for radiological image quality control management according to claim 1, characterized in that, Step S5 includes: S51. For the image quality control task, perform cross-statistical analysis according to at least the following three dimensions: Based on the dimension of the examining technician, including counting the distribution of quality control deduction items, the proportion of excellent, good, and poor evaluations, and the high-frequency problem factors for different technicians; Based on the dimension of the examination category, including comparing the differences in the qualified rates of quality control for different imaging examination types such as CT, MR, CR, and RF; Based on the dimension of the quality control factor, including analyzing the top 5 quality control defect factors with the highest occurrence frequencies in each group of examination sites; S52. For the report quality control task, perform correlation analysis according to at least the following three dimensions: Based on the dimension of the reporting doctor, including counting the distribution of the report standardization scores and clustering of error types for doctors at different professional title levels; Based on the dimension of the examination site, including analyzing the defect rates of report descriptions for different anatomical sites such as the nervous system and the skeletal system; Trend analysis based on the time dimension, including tracking the dynamic change trends of key quality control indicators at the weekly / monthly granularity; S53. Generate a visual statistical report, including a multi-dimensional data pivot table, a quality control KPI compliance matrix, and a defect heat map; Among them, the multi-dimensional data pivot table supports multi-level drill-down analysis of the hospital area, equipment model, and radiology title; The quality control KPI compliance matrix shows the achievement degrees of each examination category in terms of the qualified rate of image quality and the report integrity index; The defect heat map shows the distribution density of quality control problems for different examination rooms / equipment through spatial mapping; S54. Establish a quality control traceability mechanism to automatically associate the original examination form, quality control evaluation record, and corresponding feedback from quality control experts with the systematic quality problems discovered through statistics.

7. A radiological imaging quality control management method according to claim 1, characterized in that, It also includes the following steps: S6. Construct a dynamic feedback optimization mechanism, specifically including: S61. Based on the statistical analysis results in step S5, identify the high-frequency quality control defect factors and associated dimension combinations, and generate a risk weight matrix; S62. When configuring the subsequent sampling rules, introduce an intelligent sampling optimization engine to automatically adjust the sampling ratios of each dimension according to the risk weight matrix; S63. For the examining technician or reporting doctor whose quality control qualified rate is lower than the threshold in three consecutive statistical analyses, automatically trigger the generation of a special quality control task, and force the sampling ratio of the relevant examination forms of this person in the sampling rules to be set to 100%.

8. A radiological image quality control management system, characterized in that, It includes: A task creation module for creating quality control tasks; A sampling module for multi-dimensionally sampling examination forms according to the set sampling rules; A task assignment module for screening quality control experts according to roles and assigning quality control tasks to the selected experts based on the sampling results; A quality control operation module for quality control experts to perform quality control operations on the assigned images or reports based on a quality control template and automatically classify the quality control result levels according to the preset deduction range; A statistical analysis module for performing multi-dimensional statistical analysis on the completed quality control tasks.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the method steps described in any one of claims 1 to 7 are implemented.

10. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, the method steps described in any one of claims 1 to 7 are implemented.

Images