Cognitive function evaluation system and method based on clock drawing test and large model

Through the combination of smart terminals and servers, automatic scoring and multi-task design are used to use large language models to solve the problems of insufficient scoring objectivity and single task design in the existing technology, and comprehensive evaluation and dynamic tracking of cognitive functions are realized, improving The degree of automation and scalability of the system.

CN118969289AInactive Publication Date: 2024-11-15WENZHOU MEDICAL UNIV

Patent Information

Application Number
CN202411443493.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-16
Publication Date
2024-11-15
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing cognitive function evaluation system based on painting clock tests has problems such as insufficient objectiveness and consistency in scores, single task design, insufficient digitalization and automation, and lack of dynamic assessment and long-term tracking capabilities.

Method used

Using intelligent terminals and servers, including data acquisition module, task management module, image processing module, large language model scoring module and result analysis module, automatic scoring through large language models, multi-level and multi-task drawing clock tests are designed to achieve the The full process of data collection and analysis scores are automated, and multiple test data comparisons and trend analysis are supported.

Benefits of technology

Improve the objectivity and consistency of scores, comprehensively evaluate the subjects' cognitive ability, realize the full process automation from data collection to analytical scores, supports dynamic evaluation and long-term tracking, improves testing efficiency and scalability, and supports Comparison of results across institutions and times.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118969289A_ABST
    Figure CN118969289A_ABST
Patent Text Reader

Abstract

The invention discloses a cognitive function evaluation system and method based on a clock drawing test and a large model, the system comprises a server, the server comprises a data acquisition module, a task management module, an image processing module, a large language model scoring module and a result analysis module, the data acquisition module acquires a clock handwriting image; the task management module dynamically adjusts the difficulty and the appearance sequence of a plurality of clock drawing test tasks according to the drawing performance of the subject; the image processing module carries out image processing on the clock handwriting image; the large language model scoring module extracts image features from the processed data, analyzes the image features and the context of the clock drawing test task, and outputs a scoring result and an analysis report; and the result analysis module is used for comprehensively analyzing to obtain a cognitive function report. According to the system, multi-level and multi-task clock drawing tests are automatically scored by introducing a large language model scoring module, various cognitive competence of a subject is comprehensively evaluated, and the objectivity, consistency and reliability of scoring are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and in particular to a cognitive function assessment system and method based on a clock drawing test and a large model. Background Art

[0002] The Clock Drawing Test (CDT) is a widely used neuropsychological examination method for cognitive function assessment. The test was first proposed by Critchley in 1953 and has been widely used and studied in the following decades. CDT requires the subject to draw a complete clock in a pre-drawn circle, including all numbers and hands, and usually requires the time to be set to "11:10".

[0003] CDT can assess the functions of multiple cognitive domains, including visual-spatial ability, executive function, attention, abstract thinking, and planning and organizational ability. Due to its simplicity, rapidity, and non-invasiveness, CDT is widely used in clinical practice to screen for cognitive disorders such as Alzheimer's disease and vascular dementia.

[0004] In the existing technology, some people have developed CDT systems based on digital platforms. For example, Davis et al. developed a digital CDT system based on a tablet computer. The system uses touch screen technology to allow subjects to draw a clock directly on the tablet computer. The system can record the time parameters and handwriting characteristics during the drawing process. Scoring still relies on manual work, but the system provides some auxiliary analysis tools, such as automatic calculation of digital position deviation. Its main features: digitally record the drawing process, provide basic quantitative analysis, retain the single task mode of traditional CDT, and scoring still relies mainly on manual work. Another example: Kim et al. proposed a multimodal assessment system combining CDT and other cognitive tests. The system implements CDT on a digital platform and combines it with the Mini-Mental State Examination (MMSE) and verbal fluency test. The system uses machine learning algorithms to conduct a comprehensive analysis of multiple test results to improve the detection accuracy of cognitive impairment. Main features: combining multiple cognitive tests, using machine learning for comprehensive analysis, the CDT part still maintains the traditional single task mode, and improves the accuracy of the overall assessment, but increases the complexity and time of the test.

[0005] The main disadvantages of the above prior art are:

[0006] 1. Insufficient objectivity and consistency in scoring: Although the existing digital CDT system has improved the data collection method, the scoring still relies mainly on manual work. This makes the scoring results susceptible to the experience and subjective judgment of the assessor, reducing the reliability and consistency of the test results. The difference in scoring between different assessors may make it difficult to directly compare the test results of the same patient in different medical institutions or at different times.

[0007] 2. Single task design makes it difficult to comprehensively evaluate cognitive abilities: Most existing systems still retain the single task mode of traditional CDT. This design makes it difficult to comprehensively evaluate the various cognitive abilities of subjects, especially in distinguishing between mild cognitive impairment and normal cognitive function. A single task is difficult to effectively evaluate multiple cognitive abilities such as executive function, attention allocation, and task switching, which may lead to the neglect of certain cognitive deficits.

[0008] 3. Insufficient digitization and automation: Although some systems have achieved digital data collection, the degree of automation in data analysis and scoring is still insufficient. This not only increases the workload of medical staff, but also limits the application of the system in large-scale screening. The lack of automated analysis also makes it difficult to conduct rapid and standardized evaluations, affecting the efficiency and scalability of the test.

[0009] 4. Lack of dynamic evaluation and long-term tracking capabilities: Most existing systems focus on single test results and lack effective dynamic evaluation and long-term tracking mechanisms. This makes it difficult to monitor the changing trends of patients' cognitive functions, affecting early intervention and treatment effect evaluation. The lack of continuous data also limits in-depth research on the process of cognitive function deterioration and the formulation of personalized intervention plans. Summary of the invention

[0010] In view of the defects in the prior art, the embodiments of the present invention provide a cognitive function assessment system and method based on a clock drawing test and a large model to solve the technical problems existing in the prior art.

[0011] In a first aspect, an embodiment of the present invention provides a cognitive function assessment system based on a clock drawing test and a large model, comprising an intelligent terminal, an electromagnetic induction pen and a server, wherein the server comprises: a data acquisition module, a task management module, an image processing module, a large language model scoring module and a result analysis module; The data acquisition module is used to acquire the clock handwriting image drawn by the user according to the clock drawing test task; The task management module is used to dynamically adjust the difficulty and order of occurrence of multiple clock drawing test tasks according to the subject's drawing performance; The image processing module is used to process the clock handwriting image to obtain processed data; The large language model scoring module is used to extract image features from the processed data, analyze the image features and the context of the clock drawing test task, determine the subject's completion of the clock drawing test task, and output a scoring result and an analysis report; The result analysis module is used to comprehensively analyze the scoring results and analysis reports of multiple tasks to obtain a cognitive function report.

[0012] In a second aspect, an embodiment of the present invention provides a method for evaluating cognitive function based on a clock drawing test and a large model, comprising: Obtaining a clock handwriting image drawn by a user according to a clock drawing test task; The difficulty and order of presentation of multiple clock drawing test tasks were dynamically adjusted according to the subjects’ drawing performance; Performing image processing on the clock handwriting image to obtain processed data; Extract image features from the processed data, analyze the image features and the context of the clock drawing test task, determine the subject's completion of the clock drawing test task, and output the scoring results and analysis report; The scoring results and analysis reports of multiple tasks are comprehensively analyzed to obtain a cognitive function report.

[0013] Beneficial effects of the present invention:

[0014] The embodiment of the present invention provides a cognitive function assessment system based on a clock drawing test and a large model. By introducing a large language model scoring module for automatic scoring, the influence of human factors is reduced, the objectivity and consistency of scoring are improved, the reliability of scoring results is ensured, and cross-institutional and cross-time result comparison is supported.

[0015] By designing a multi-level, multi-task clock drawing test, we can comprehensively evaluate the subjects' various cognitive abilities, including visual-spatial ability, executive function, attention, abstract thinking, etc., thereby improving the system's ability to distinguish different degrees of cognitive impairment, especially the detection sensitivity for mild cognitive impairment.

[0016] The system automates the entire process from data collection to analysis and scoring, improves test efficiency, and supports large-scale screening applications. Automated analysis greatly reduces manual workload and improves the standardization and repeatability of evaluation.

[0017] The system supports multiple test data comparison and trend analysis functions, realizing dynamic evaluation and long-term tracking of cognitive function. The system creates a unique digital file for each user, recording the detailed results and scores of each test. The result analysis module can generate a graphical long-term trend analysis to intuitively display the changes in cognitive function. It helps to timely discover subtle changes in cognitive function and supports early intervention and the formulation of personalized treatment plans.

[0018] A cognitive function assessment method based on a clock drawing test and a large model provided in an embodiment of the present invention and a cognitive function assessment system based on a clock drawing test and a large model are based on the same inventive concept and have the same beneficial effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0020] Figure 1 A structural block diagram of a cognitive function assessment system based on a clock drawing test and a large model provided by a first embodiment of the present invention is shown;

[0021] Figure 2 A flowchart of a cognitive function assessment method based on a clock drawing test and a large model provided by another embodiment of the present invention is shown. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0024] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0025] It should be further understood that the term "and / or" used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0026] As used in this specification and the appended claims, the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrases "if it is determined" or "if [described condition or event] is detected" may be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0027] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by those skilled in the art to which the present invention belongs.

[0028] like Figure 1 As shown, a structural block diagram of a cognitive function assessment system based on a clock drawing test and a large model provided by a first embodiment of the present invention is shown, the system comprising: an intelligent terminal, an electromagnetic induction pen and a server, the server comprising: a data acquisition module, a task management module, an image processing module, a large language model scoring module and a result analysis module; the data acquisition module is used to obtain a clock handwriting image drawn by a user according to a clock drawing test task; the task management module is used to dynamically adjust the difficulty and appearance order of multiple clock drawing test tasks according to the subject's drawing performance; the image processing module is used to perform image processing on the clock handwriting image to obtain processed data; the large language model scoring module is used to extract image features from the processed data, analyze the image features and the context of the clock drawing test task, determine the subject's completion of the clock drawing test task, and output a scoring result and an analysis report; the result analysis module is used to comprehensively analyze the scoring results and analysis reports of multiple tasks to obtain a cognitive function report.

[0029] The cognitive function assessment system based on the clock drawing test and the large model includes a hardware system intelligent terminal and an electromagnetic induction pen. The intelligent terminal can be a smart phone and a tablet computer. In this embodiment, a high-resolution touch screen tablet computer with a size of 10.5 inches or more is used, and a device with an electromagnetic screen function is preferred. Electromagnetic screen technology can accurately capture handwriting features such as pressure and tilt angle, providing richer data for subsequent analysis. The tablet computer is equipped with a high-performance processor and sufficient memory to support real-time image processing and LLM operations. Electromagnetic induction pen: A high-precision electromagnetic induction pen used in conjunction with a touch screen can simulate a real writing experience and provide 1024 levels or higher pressure sensing. The system uses a high-precision touch screen tablet computer as a hardware platform, and with an electromagnetic induction pen, it can accurately capture the user's painting process.

[0030] The server includes a user interface module, a data acquisition module, a task management module, an image processing module, a large language model scoring module, a result analysis module, and a calibration module. The user interface module uses an intuitive graphical user interface (GUI) with large fonts and high contrast, which is suitable for users of all ages. The interface includes a task instruction area, a drawing area, and a function button area. Task instructions use both text and voice prompts to ensure that the subjects fully understand the task requirements. Subjects can complete the drawing task directly on the user interface according to the task instructions using an electromagnetic pen, or by taking pictures of paper drawings. The system records handwriting data in real time, including position, pressure, and timestamps.

[0031] This embodiment designs four progressive test tasks to comprehensively evaluate the cognitive function of the subject:

[0032] 1. Task 1: Standard CDT Description: Draw a complete clock face in a pre-drawn circle and set the time to "11:10". Implementation: The system displays a circle with a diameter of 60% of the screen width. The user draws inside the circle using an electromagnetic pen. The system captures the handwriting in real time and analyzes it at the end of the task. Assessment: Overall cognitive function, visual-spatial ability, number recognition and sequencing ability.

[0033] 2. Task 2: Dynamic time calculation Description: Given a base time (e.g. "3:15"), you are required to draw the time 2 hours and 30 minutes later. Implementation: The system randomly generates a base time and a time offset (±1 to ±4 hours). The user needs to mentally calculate and draw the result. Assessment: Mathematical ability, abstract thinking, working memory.

[0034] 3. Task 3: Reverse Clock Description: Draw a clock that runs counterclockwise and the time is set to "2:45". Implementation: The system provides a blank drawing area and the user needs to draw the entire clock from scratch. LLM evaluates the reverse order of the number arrangement and the correctness of the pointer position. Assessment: Cognitive flexibility, inhibitory control, spatial thinking ability.

[0035] 4. Task 4: Mirror clock Description: Draw a mirror reflection of a given time ("9:20"). Implementation: The system displays a normal clock image and the user draws a mirror version on the opposite side. LLM analyzes the symmetry and accuracy of the drawing. Assessment: Visuospatial ability, mental rotation ability, executive function.

[0036] Compared with the single task mode commonly used in the prior art, the present invention designs four hierarchical tasks to achieve a comprehensive assessment of the cognitive ability of the subject. The four tasks are: standard CDT, dynamic time calculation, reverse clock and mirror clock, each task targets a specific cognitive area. For example, the dynamic time calculation task assesses mathematical ability and abstract thinking, the reverse clock task tests cognitive flexibility and inhibitory control, and the mirror clock task examines visual spatial ability and mental rotation ability. This multi-task design significantly improves the system's ability to distinguish different degrees of cognitive impairment, especially showing higher sensitivity in detecting mild cognitive impairment.

[0037] By comprehensively analyzing the performance of these tasks, the system can generate a comprehensive cognitive function report, including the scores of various cognitive abilities, relative strengths and weaknesses, and comparison with the normal range of age matching. This provides clinicians with richer and more accurate diagnostic evidence and helps to develop more targeted intervention strategies.

[0038] The data acquisition module of the embodiment of the present invention is used to obtain the clock handwriting image drawn by the user according to the clock drawing test task. The task management module is used to control the presentation and switching of four hierarchical tasks. The task management module is based on the state machine design and can dynamically adjust the difficulty and order of tasks according to the performance of the subject. For example, if the subject performs well in the standard CDT task, the system will automatically increase the difficulty of subsequent tasks.

[0039] The image processing module is used to process the clock handwriting image to obtain the processed data. The specific image processing process is: a. Convert the handwriting data into a high-resolution (such as 2048x2048 pixels) grayscale image; b. Apply image enhancement technology, such as adaptive thresholding, to improve line clarity; c. Use morphological operations (such as dilation and erosion) to optimize image quality.

[0040] The Large Language Model (LLM) scoring module is a scoring system built on a pre-trained large language model (such as GPT-4 or its customized version). It is used to extract image features from processed data, analyze the image features and the context of the clock drawing test task, determine the subject's completion of the clock drawing test task, and output the scoring results and analysis report. The scoring criteria are based on Freedman's 15-point system or Rouleau's 10-point system, but are expanded to adapt to multi-task design. The large language model scoring module includes a scoring unit, which scores the integrity and symmetry of the drawn circle, the correctness, position and arrangement of the numbers, the accuracy and proportion of the pointer, the overall layout and space utilization, and the completion of task-specific requirements (such as reverse clocks and mirror clocks) during the scoring process to obtain scoring results and analysis reports.

[0041] The result analysis module is used to comprehensively analyze the scoring results and analysis reports of multiple tasks to obtain a cognitive function report. The report includes: scores and percentile rankings of various cognitive abilities, relative strength and weakness analysis, comparison with the normal range matched to age, suggestions for improving cognitive function, and long-term trend charts.

[0042] The calibration module is used to regularly compare the scoring results output by the large language model scoring module with the scoring results of professional doctors to maintain scoring consistency. The calibration process is as follows: a. Collect a batch of samples scored by professional doctors (at least 100 copies); b. Use these samples to fine-tune the LLM scoring model; c. Calculate the consistency index between the LLM score and the doctor's score (such as Cohen's Kappa coefficient); d. If the consistency is lower than the preset threshold, the system will prompt that manual review and adjustment are required.

[0043] The user interface module is also used to display brief results and generate detailed PDF reports, which include charts and explanations. If an abnormality is detected, a doctor notification is generated and sent to the doctor's terminal.

[0044] The server of the embodiment of the present invention also includes a data management module for managing and storing test data, creating a file for each user, recording the detailed results and scores of each test of the user, and encrypting, backing up and version controlling the test data.

[0045] The cognitive function assessment system based on the clock drawing test and the large model provided by the embodiment of the present invention has the following advantages:

[0046] 1. Improved objectivity and consistency of scoring: By introducing an automatic scoring system based on a large language model (LLM), the objectivity and consistency of scoring are significantly improved. Unlike existing technologies that mainly rely on manual scoring, this system uses advanced computer vision technology and LLM for automatic analysis and scoring, effectively reducing the impact of human factors.

[0047] Specifically, the system's image processing module converts user drawings into standardized digital images, and then uses a convolutional neural network to extract key features. These features are then input into a specially trained large language model scoring module for in-depth analysis and scoring. This process ensures consistency and repeatability of the scoring criteria, and highly consistent results can be obtained regardless of when and where the test is conducted.

[0048] In addition, the system also includes a calibration function, which regularly compares and adjusts the scoring results with those of professional doctors, further ensuring the accuracy of the scoring and its consistency with clinical practice. This not only improves the reliability of a single test, but also supports the comparison of results across institutions and time, providing a more reliable data basis for long-term research and clinical practice.

[0049] 2. Achieved comprehensive cognitive ability assessment: Compared with the single task mode commonly used in the prior art, the embodiment of the present invention designed four hierarchical tasks to achieve a comprehensive assessment of the cognitive ability of the subject. The four tasks are: standard CDT, dynamic time calculation, reverse clock and mirror clock, each of which targets a specific cognitive field. By comprehensively analyzing the performance of these tasks, the system can generate a comprehensive cognitive function report, including scores for various cognitive abilities, relative strengths and weaknesses analysis, and comparison with age-matched normal ranges. This provides clinicians with a richer and more accurate basis for diagnosis, and helps to formulate more targeted intervention strategies.

[0050] 3. Improved digitization and automation of the system: The embodiment of the present invention realizes full process automation from data collection to analysis and scoring, significantly improving test efficiency and scalability. The system uses a high-precision touch screen tablet computer as the hardware platform, and with an electromagnetic induction pen, it can accurately capture the user's painting process. The software system automatically processes the collected data and completes the entire evaluation process from image processing, feature extraction to LLM analysis without human intervention. This highly automated design not only greatly reduces the manual workload, but also improves the standardization and repeatability of the evaluation. The system can quickly process large amounts of data and support large-scale screening applications, which is of great significance in the screening of early cognitive impairment in the context of an aging population.

[0051] In addition, the system's data management module realizes the secure storage and management of test data, supports data encryption, backup and version control, and provides a reliable data foundation for long-term research and clinical applications.

[0052] 4. Dynamic evaluation and long-term tracking capabilities are realized: An important innovation of the embodiment of the present invention is the realization of dynamic evaluation and long-term tracking of cognitive function. The system creates a unique digital file for each user, recording the detailed results and scores of each test. The result analysis module can generate a graphical long-term trend analysis to intuitively display the changes in cognitive function.

[0053] This dynamic assessment capability enables the system to detect subtle changes in cognitive function in a timely manner, providing an important basis for early intervention. For example, the system can detect a slight decline in a certain cognitive ability, which can provide early warnings to doctors and patients even if the decline has not yet reached the clinical diagnostic standard. The long-term tracking function also supports the objective evaluation of treatment effects. Doctors can accurately evaluate the effectiveness of treatment plans by comparing test results before and after intervention, and make adjustments as needed. This provides data support for the formulation and optimization of personalized treatment plans.

[0054] In the first embodiment described above, a cognitive function assessment system based on a clock drawing test and a large model is provided. Correspondingly, the present application also provides a cognitive function assessment method based on a clock drawing test and a large model. Figure 2 , which is a flow chart of a cognitive function assessment method based on a clock drawing test and a large model provided by the second embodiment of the present invention. Since the method embodiment is basically similar to the device embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the device embodiment. The method embodiment described below is only illustrative.

[0055] like Figure 2 , which shows a flow chart of a cognitive function assessment method based on a clock drawing test and a large model provided by an embodiment of the present invention, including: Obtaining a clock handwriting image drawn by a user according to a clock drawing test task; The difficulty and order of presentation of multiple clock drawing test tasks were dynamically adjusted according to the subjects’ drawing performance; Performing image processing on the clock handwriting image to obtain processed data; Extract image features from the processed data, analyze the image features and the context of the clock drawing test task, determine the subject's completion of the clock drawing test task, and output the scoring results and analysis report; The scoring results and analysis reports of multiple tasks are comprehensively analyzed to obtain a cognitive function report.

[0056] Among them, the clock drawing test tasks include standard CDT tasks, dynamic time calculation tasks, direction clock tasks, and mirror clock tasks. The specific methods for regularly comparing the scoring results output by the large language model scoring module with the scoring results of professional doctors include: Obtain samples scored by professional doctors and use them to fine-tune the large language scoring model; The consistency index between the scores of the large language scoring model and the scores of the physicians was calculated; The consistency index is compared with the preset threshold. If the consistency index is lower than the preset threshold, a prompt message is issued indicating that manual review and adjustment are required.

[0057] The method also includes: regularly comparing the scoring results output by the large language model scoring module with the scoring results of professional doctors to maintain scoring consistency.

[0058] The specific method of analyzing the image features and the context of the clock drawing test task to determine whether the subject has completed the clock drawing test task includes: Scores are given based on the completeness and symmetry of the drawn circle, the correctness of the numbers, the position and arrangement, the accuracy and proportion of the pointers, the overall layout and space utilization, and the completion of the task, and a scoring result and analysis report are obtained.

[0059] A cognitive function assessment method based on a clock drawing test and a large model and a cognitive function assessment system based on a clock drawing test and a large model provided in an embodiment of the present invention are based on the same inventive concept and have the same beneficial effects, which will not be described in detail herein.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A cognitive function assessment system based on clock drawing test and large model, including an intelligent terminal, an electromagnetic induction pen and a server, It is characterized in that The server includes: a data acquisition module, a task management module, an image processing module, a large language model scoring module and a result analysis module; The data acquisition module is used to acquire the clock handwriting image drawn by the user according to the clock drawing test task; The task management module is used to dynamically adjust the difficulty and order of occurrence of multiple clock drawing test tasks according to the subject's drawing performance; The image processing module is used to process the clock handwriting image to obtain processed data; The large language model scoring module is used to extract image features from the processed data, analyze the image features and the context of the clock drawing test task, determine the subject's completion of the clock drawing test task, and output a scoring result and an analysis report; The result analysis module is used to comprehensively analyze the scoring results and analysis reports of multiple tasks to obtain a cognitive function report.

2. The system according to claim 1, It is characterized in that The system also includes a calibration module, which is used to regularly compare the scoring results output by the large language model scoring module with the scoring results of professional doctors to maintain scoring consistency.

3. The system according to claim 2, It is characterized in that The calibration module includes a comparison unit, which is used to collect samples scored by professional doctors, use the samples to fine-tune the large language model scoring module, and calculate the consistency index between the score of the large language model scoring module and the doctor's score. If the consistency index is lower than a preset threshold, a prompt message requiring manual review and adjustment is issued.

4. The system according to claim 1, It is characterized in that The clock drawing test tasks include a standard CDT task, a dynamic time calculation task, a direction clock task and a mirror clock task.

5. The system according to claim 4, It is characterized in that The large language model scoring module includes a scoring unit, which scores according to the integrity and symmetry of the drawn circle, the correctness, position and arrangement of the numbers, the accuracy and proportion of the pointer, the overall layout and space utilization, and the completion of the task during the scoring process to obtain a scoring result and an analysis report.

6. A method for evaluating cognitive function based on the clock drawing test and the large model, It is characterized in that include: Obtaining a clock handwriting image drawn by a user according to a clock drawing test task; The difficulty and order of presentation of multiple clock drawing test tasks were dynamically adjusted according to the subjects’ drawing performance; Performing image processing on the clock handwriting image to obtain processed data; Extract image features from the processed data, analyze the image features and the context of the clock drawing test task, determine the subject's completion of the clock drawing test task, and output the scoring results and analysis report; The scoring results and analysis reports of multiple tasks are comprehensively analyzed to obtain a cognitive function report.

7. The method according to claim 5, It is characterized in that Also includes: Regularly compare the scoring results output by the large language model scoring module with the scoring results of professional doctors to maintain scoring consistency.

8. The method according to claim 7, It is characterized in that The specific method of regularly comparing the scoring results output by the large language model scoring module with the scoring results of professional doctors includes: Obtain samples scored by professional doctors and use them to fine-tune the large language scoring model; The consistency index between the scores of the large language scoring model and the scores of the physicians was calculated; The consistency index is compared with the preset threshold. If the consistency index is lower than the preset threshold, a prompt message is issued indicating that manual review and adjustment are required.

9. The method according to claim 6, It is characterized in that The clock drawing test tasks include a standard CDT task, a dynamic time calculation task, a direction clock task and a mirror clock task.

10. The method according to claim 9, It is characterized in that The specific method of analyzing the image features and the context of the clock drawing test task to determine whether the subject has completed the clock drawing test task includes: Scores are given based on the completeness and symmetry of the drawn circle, the correctness of the numbers, the position and arrangement, the accuracy and proportion of the pointers, the overall layout and space utilization, and the completion of the task, and a scoring result and analysis report are obtained.

Citation Information

Patent Citations

  • Cognition impairment evaluating system and method based on clock drawing test

    CN104715157A

  • Adaptive cognitive function assessment method and system

    CN108133736A

  • Clock drawing test assessment system and method

    CN108197564A

  • Touch interactive motion user feature extraction method based on general drawing task and auxiliary disease detection system

    CN111354458A

  • Method and device for quickly screening cognitive disorder of old people in hierarchical progressive mode, processor and computer storage medium thereof

    CN113827191A

Cited By

  • Device and method for evaluating high-order cognition and emotion of children based on drawing pixel time sequence information

    CN120419957A