Universal cognitive evaluation system

Abstract

The invention provides a universal cognitive assessment system, comprising a patient management module used for storing and managing identity information and test related information of a patient; the task test module is used for providing various test tasks for evaluating cognitive competence for the to-be-tested patient through a touch device, and acquiring test feedback data of the to-be-tested patient through a touch screen and / or a camera; and the task report module is used for evaluating the cognitive ability of the to-be-tested patient through a preset neural network-based cognitive evaluation model according to the test feedback data of the to-be-tested patient so as to generate a cognitive evaluation report of the to-be-tested patient.

Description

Technical Field

[0001] The present invention relates to the field of cognitive assessment, in particular to the field of cognitive assessment based on artificial intelligence, and more particularly to a universal cognitive assessment system. Background Art

[0002] With the increase in life expectancy and the decline in fertility, the world is facing the problem of an aging population. According to the United Nations World Population Prospects, the proportion of people aged 65 or above will increase from 10% in 2022 to 16% in 2050. At the same time, the number of people suffering from neurodegenerative diseases such as Alzheimer's disease (AD) and mild cognitive impairment (MCI) has increased significantly. When the disease is severe, it will greatly affect people's daily lives. Therefore, it is very necessary to conduct cognitive assessments on the elderly on a regular basis.

[0003] Traditional cognitive assessments, such as the Mini-Mental State Examiniation (MMSE) and the Montreal Cognitive Assessment (MoCA), not only require a lot of human resources, but may also affect the accuracy of the assessment results due to excessive tension of the patient or changes in the environment, and have a certain degree of subjectivity. Moreover, the process of manually recording and analyzing assessment data is complicated and prone to errors, making it difficult to efficiently achieve early screening and monitoring of AD. In addition, laboratory tests and imaging examinations require relatively high professional skills and costs, and are not universal. With the continuous development of artificial intelligence technology, many intelligent tools for assessing patients' cognitive conditions have emerged. Although these tools have simplified the testing process to a certain extent, most of them can only provide basic test results, lacking in-depth analysis of the data and detailed reporting functions. This limitation makes it difficult for these tools to fully reflect the patient's cognitive status in actual applications, and cannot provide effective decision support and personalized treatment recommendations for medical staff.

[0004] In the field of cognitive assessment, existing technologies have the following problems:

[0005] Time-consuming: Traditional cognitive assessment methods usually require professional medical staff to test and record patients face to face. The whole process is time-consuming and it is difficult to complete a large amount of screening work in a short period of time.

[0006] Not objective: Existing assessment methods often rely on the observation and judgment of medical staff, which are easily affected by subjective factors, resulting in the assessment results being not objective and consistent;

[0007] Professional limitations: Traditional assessment methods require professional equipment and well-trained personnel, which increases the cost of assessment and limits its application in communities or families with limited resources;

[0008] No detailed analysis: Many existing assessment tools only provide simple test results and lack in-depth data analysis and detailed reporting functions, which cannot provide effective decision support and personalized treatment recommendations for medical staff;

[0009] High cost: Although some advanced assessment systems have automated functions, they usually rely on expensive equipment and complex operating procedures, which reduces their convenience and popularity in clinical practice.

[0010] It should be noted that this background technology is only used to introduce the relevant information of the present invention to help understand the technical solution of the present invention, but it does not mean that the relevant information is necessarily the prior art. The relevant information is submitted and disclosed together with the present invention solution. If there is no evidence that the relevant information has been disclosed before the application date of the present invention, the relevant information shall not be regarded as the prior art. Summary of the invention

[0011] Therefore, the purpose of the present invention is to overcome the above-mentioned defects of the prior art and provide a universal cognitive assessment system.

[0012] The objective of the present invention is achieved through the following technical solutions:

[0013] According to a first aspect of the present invention, there is provided a cognitive assessment system, comprising a patient management module for storing and managing identity information and test-related information of a patient; a task testing module for providing a variety of test tasks for assessing cognitive ability to a patient to be tested via a touch device, and obtaining test feedback data of the patient to be tested via a touch screen and / or a camera; and a task reporting module for assessing the cognitive ability of the patient to be tested via a preset neural network-based cognitive assessment model according to the test feedback data of the patient to be tested, so as to generate a cognitive assessment report for the patient to be tested.

[0014] Optionally, the cognitive assessment report includes scores of multiple cognitive domains and a comprehensive score of cognitive ability, wherein the cognitive assessment model determines the scores of the multiple cognitive domains based on the test feedback data, and determines the comprehensive score of cognitive ability based on the scores of the multiple cognitive domains.

[0015] Optionally, the scores of the multiple cognitive domains include scores of flexibility, memory, attention, perception and agility, or a combination thereof.

[0016] Optionally, the cognitive assessment model includes: a designated area feature extraction module, which is used to extract a first feature of a set designated area from the image data of the test feedback data using a neural network; a designated area identification module, which is used to determine the identification result of the designated area based on the first feature of each designated area; an automatic feature extraction module, which is used to automatically extract a second feature from the image data and associated data of the test feedback data using a neural network without setting a designated area, wherein the associated data includes the test time and / or the test accuracy; a feature fusion module, which is used to fuse the identification result of the designated area with the second feature to obtain a comprehensive feature; a first evaluation module, which is used to evaluate the scores of the patient to be tested in a preset plurality of cognitive domains based on the comprehensive features; and a second evaluation module, which is used to determine a comprehensive score of cognitive ability based on the scores of the plurality of cognitive domains.

[0017] Optionally, the cognitive assessment model includes: a designated area feature extraction module, which is used to extract a first feature of a set designated area from the image data of the test feedback data using a neural network; a designated area identification module, which is used to determine the identification result of the designated area based on the first feature of each designated area; an automatic feature extraction module, which is used to automatically extract a second feature from the image data and associated data of the test feedback data using a neural network without setting a designated area, and to extract the causal relationship between each brain area from the brain sensor data, wherein the associated data includes the test time and / or the test accuracy, and the brain sensor data is collected from the patient to be tested during the test process; a feature fusion module, which is used to fuse the identification result of the designated area, the second feature and the causal relationship to obtain a comprehensive feature; a first evaluation module, which is used to evaluate the scores of the patient to be tested in a preset plurality of cognitive domains based on the comprehensive features; and a second evaluation module, which is used to determine the comprehensive score of cognitive ability based on the scores of the plurality of cognitive domains.

[0018] Optionally, the cognitive assessment model is trained in the following manner: a training set is obtained, which includes multiple samples, each sample includes sample person data and a label, the sample person data includes the sample person's test feedback data and brain sensor data, and the label includes the sample person's cognitive ability comprehensive score reference value and the score reference values ​​of multiple cognitive domains; using the training set and a preset loss function, the cognitive assessment model is trained to evaluate the sample person's cognitive ability comprehensive score and the scores of multiple cognitive domains, wherein minimizing the loss function value is used as the optimization goal, and the parameters of the cognitive assessment model are updated according to the evaluated cognitive ability comprehensive score and the scores of multiple cognitive domains and the sample person's cognitive ability comprehensive score reference value and the score reference value of multiple cognitive domains recorded in the corresponding labels.

[0019] Optionally, the multiple test tasks include the following test tasks or a combination of the following test tasks: a box and block test task, which is used to provide a box containing multiple blocks, the interior of the box is divided into at least two areas, and the patient is instructed to move the blocks in one area to another area within a specified time; a line connection test, which is used to provide multiple connection points, and the patient is instructed to connect the connection points according to preset requirements; a clock drawing test, which is used to instruct the patient to draw a clock face with scales and hour time numbers, and instruct the patient to draw an hour hand and a minute hand corresponding to a specified time on the clock face; a pentagon drawing test, which is used to instruct the patient to draw two partially overlapping pentagons as required, and the overlapping parts form a preset shape; a complex figure test, which is used to provide a complex figure to be copied, instruct the patient to observe and copy the complex figure, and instruct the patient to redraw the complex figure by recalling after various time lengths.

[0020] Optionally, the brain sensing data includes electroencephalogram (EEG) and / or functional near infrared spectroscopy (FNIRS) data of the brain.

[0021] Optionally, the system further includes: a doctor management module for managing doctor registration information; wherein the system is configured so that doctors can add patients and select test tasks for which patients need to participate.

[0022] According to a second aspect of the present invention, a touch control device is provided, in which the system as described in the first aspect is deployed, and the touch control device is a device with a touch control screen and a camera.

[0023] Optionally, the touch device is a mobile phone, a tablet computer, or a desktop computer with a touch display. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The embodiments of the present invention are further described below with reference to the accompanying drawings, in which:

[0025] Figure 1 A schematic diagram of a system framework according to an embodiment of the present invention;

[0026] Figure 2 is a schematic diagram of system modules according to an embodiment of the present invention;

[0027] Figure 3 A schematic diagram of the overall flow of a system according to an embodiment of the present invention;

[0028] Figure 4 A schematic diagram of a working process of a testing phase according to an embodiment of the present invention;

[0029] Figure 5 Schematic diagram of the work flow of the cognitive assessment stage according to an embodiment of the present invention;

[0030] Figure 6A schematic diagram of the main interface of the system according to an embodiment of the present invention;

[0031] Figure 7 A schematic diagram of a registration interface according to an embodiment of the present invention;

[0032] Figure 8 A schematic diagram of an interface for filling in doctor information according to an embodiment of the present invention;

[0033] Fig. 9 A schematic diagram of a login interface according to an embodiment of the present invention;

[0034] Fig.10 A schematic diagram of a patient management interface according to an embodiment of the present invention;

[0035] Fig.11 A schematic diagram of a task testing process according to an embodiment of the present invention;

[0036] Fig.12 A schematic diagram for selecting a task according to an embodiment of the present invention;

[0037] Fig.13 A schematic diagram of a subtask list according to an embodiment of the present invention;

[0038] Fig.14 is a schematic diagram of a cognitive report interface according to an embodiment of the present invention;

[0039] Fig.15 Schematic diagram of a doctor information management interface according to an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below through specific embodiments in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0041] As mentioned in the background technology section, the existing technology has problems such as long time consumption, objectivity, professional limitations, no detailed analysis and high cost. These problems and shortcomings are caused by the technical limitations of traditional evaluation methods. For example, the problems of long time consumption and strong subjectivity are mainly caused by manual operation and human judgment; the problems of high cost and poor convenience of use are closely related to professional equipment and complex processes; insufficient data analysis reflects the shortcomings of existing tools in data processing and report generation. The present invention uses artificial intelligence technology to develop a cognitive assessment system (called cScreen), which uses a task test module to provide a variety of test tasks for evaluating cognitive ability to the patient to be tested, wherein the test feedback of the patient to be tested is obtained through a touch screen and / or a camera during the test, and the brain sensor data of the patient to be tested during the test feedback is collected through a brain information collector; and the task report module is used to evaluate the cognitive ability of the patient to be tested through a preset neural network-based cognitive assessment model according to the identity information, test feedback and brain sensor data of the patient to be tested, so as to generate a cognitive assessment report for the patient to be tested. The system overcomes the shortcomings of traditional assessment methods, which are time-consuming, subjective and costly; it not only improves the efficiency and accuracy of diagnosis of cognitive diseases, but also has the functions of data analysis and detailed reporting, realizing simple and efficient cognitive assessment.

[0042] According to an embodiment of the present invention, a schematic system architecture diagram is shown as follows: Figure 1 As shown. First, the user of the system can be a doctor or a patient. The patient can be some people who are suspected of cognitive impairment, mild cognitive impairment or Alzheimer's disease screened by the doctor. The doctor hopes to conduct a more detailed cognitive assessment of the patient through the system. Secondly, the applicable device of the system is a touch device, such as: a large touch screen, a computer, a tablet or a mobile phone. By installing an APP (application) or a small program on the touch device, the service provided by the system can be obtained in the form of an APP or a small program. In addition, the system of the embodiment of the present invention includes a data flow path, a functional layer and an algorithm layer. In the data flow path, the transmission and storage of network requests and various data (such as test data, audio data, signal data, JSON data) are realized through interaction with the server and database (such as MySQL / Oracle). The functional layer includes six modules: login, registration, patient management, task testing, task reporting and doctor management. Through the collaborative cooperation between modules, an integrated cognitive assessment system is constructed to realize the comprehensive operation and efficient management of the cognitive assessment system, improve the assessment quality and patient medical experience. The algorithm layer analyzes the patient's cognitive ability through feature extraction of test feedback data of various tasks, causal discovery algorithm, and cognitive assessment module to assist doctors in making diagnostic decisions. See Figure 2 , a schematic cognitive assessment system includes the following modules:

[0043] The login module 201 is used to provide login services for users, including doctors and / or patients.

[0044] The registration module 202 is used to provide user registration services; for example, providing a registration interface and guidance.

[0045] The doctor management module 203 is used to manage the doctor's registration information; for example, to store and manage the user ID (name or user name), mobile phone number and login password, etc.

[0046] The patient management module 204 is used to store and manage the patient's identity information and test-related information; for example, to store and manage the patient's age, gender and / or patient condition description; test-related information includes historical comprehensive evaluation reports and / or test feedback results.

[0047] The task test module 205 is used to provide the patient to be tested with a variety of test tasks for evaluating cognitive ability through a touch device and / or obtain the brain sensor data of the patient to be tested during the test feedback through a brain information collector. Among them, an optional method for obtaining the test feedback data of the patient to be tested is provided before the test, and the optional methods include a touch screen and a camera. For example, the doctor can choose to input the test feedback data of a certain test task by operating the touch screen or taking a photo with a camera according to the patient's condition.

[0048] The task reporting module 206 is used to evaluate the cognitive ability of the patient to be tested through a preset neural network-based cognitive assessment model according to the test feedback data of the patient to be tested, so as to generate a cognitive assessment report for the patient to be tested. The cognitive assessment report can be divided into a cognitive assessment comprehensive report, an assessment report for each cognitive domain, and an assessment result for each task. The cognitive assessment comprehensive report includes a description of the patient's overall situation, including a comprehensive score of cognitive ability and scores for multiple cognitive domains. A single cognitive domain assessment report includes the patient's score in the cognitive domain and a description of the patient's situation in the cognitive domain. Multiple cognitive domains include, for example, flexibility, memory, attention, perception, and agility or a combination thereof. A single task assessment result includes a description of the patient's score on a single test task and / or a test performance on the test task.

[0049] In order to more conveniently and efficiently evaluate the patient's cognitive ability, the present invention provides a plurality of test tasks suitable for execution on a touch screen or paper. According to one embodiment of the present invention, the plurality of test tasks include the following test tasks or a combination of the following test tasks: a box and block test task, a line connection test, a clock drawing test, a pentagon drawing test, and a complex figure test. Among them:

[0050] The Box and Block Test (BBT) is used to provide a box containing multiple blocks, which is divided into at least two areas, and instruct the patient to move the blocks in one area to another area within a specified time. For example: This test uses a touch screen tablet to simulate a wooden box, which is divided into two areas in the middle by a baffle, and multiple blocks are placed on one side of the baffle. During the test, the patient to be tested uses his / her testing hand to move as many blocks as possible from one area of ​​the box to another area of ​​the same size (1 block is moved at a time) within a specified time (60 seconds for the test and 15 seconds for the practice).

[0051] Trail Making Test (TMT) is used to provide multiple connection points and instruct the patient to connect the points according to the preset requirements. For example, during the test, circles of numbers from 1 to 25 will be presented. They are scattered and irregular. The first 1 and the last 25 are marked with the start (Begin) and the end (End) as prompts. Then they need to be connected in the order of the natural arrangement of the numbers. Numbers cannot be skipped. It is required to connect as quickly as possible while maintaining accuracy. Of course, the patient can also be instructed to answer on paper, such as printing out the test screen in advance or asking the patient to draw on the spot according to the screen display before answering.

[0052] The clock drawing test is used to instruct the patient to draw a clock face with scales and hourly time numbers, and to instruct the patient to draw the hour and minute hands corresponding to a specified time on the clock face. For example: the test requires the patient to draw a clock face and write the numbers representing the time in the correct position. After the patient draws a circle and fills in the numbers, the patient is asked to draw the minute and hour hands and point the time to 11:10. It should be understood that the patient can also be instructed to draw other specified times, such as 6:30, 10 o'clock, etc.

[0053] The pentagon drawing test is used to instruct the patient to draw two partially overlapping pentagons as required, and the overlapping parts form a preset shape. For example: the test requires the patient to copy two partially overlapping pentagons, which have 10 corners, 2 intersecting straight lines forming a quadrilateral, and the overlapping parts form a rhombus (i.e., the preset shape). Of course, different details can also be designed to form different subtasks. For example, the overlapping parts form a rhombus as a subtask, while another subtask requires the overlapping parts to form a triangle.

[0054] The complex figure test is used to provide complex figures to be copied, instruct the patient to observe and copy the complex figures, and instruct the patient to redraw the complex figures by recall after various time periods. For example: the test requires the patient to copy and draw the Rey-Osterrieth complex figure, the tester records the copying time, and removes the figure card immediately after the copying is completed; the patient to be tested redraws the figure according to the memory after an interval of about 3 minutes after the copying is completed, and the patient to be tested needs to record the instantaneous recall time again; the patient to be tested draws the figure for the third time according to the memory after an interval of about 30 minutes after the copying is completed, and the patient to be tested needs to record the delayed recall time for the third time. Each drawing should not exceed 10 minutes.

[0055] In addition, implementers may also set some additional tasks to assist the test, such as 100 countdown (i.e., counting down from 100 to 0), addition and subtraction, and / or animal naming (i.e., matching the corresponding animal name according to the animal picture shown), etc.

[0056] It should be understood that those skilled in the art may selectively deploy some of the above test tasks in the system, or even add some other types of test tasks suitable for completion on a touch screen device and / or on paper, to achieve similar effects.

[0057] In order to more efficiently and accurately evaluate the test feedback data of the patient to be tested, a cognitive assessment model based on a neural network can be set up.

[0058] According to an optional embodiment of the present invention, the cognitive assessment model includes: a designated area feature extraction module, which is used to extract the first feature of the designated area from the image data of the test feedback data using a neural network; a designated area recognition module, which is used to determine the recognition result of the designated area according to the first feature of each designated area; an automatic feature extraction module, which is used to automatically extract the second feature from the image data and associated data of the test feedback data using a neural network without setting the designated area, wherein the associated data includes the test time and the test accuracy; a feature fusion module, which is used to fuse the recognition result and the second feature of the designated area to obtain a comprehensive feature; a first evaluation module, which is used to evaluate the scores of the patient to be tested in a preset multiple cognitive domains according to the comprehensive feature; and a second evaluation module, which is used to determine the comprehensive score of cognitive ability according to the scores of the multiple cognitive domains. The designated area feature extraction module can first obtain an image block of the designated area in the image data, extract the first feature of the designated area according to the image block, and determine the recognition result of the designated area according to the first feature. The designated area can be a key area set by the implementer for extracting information, which can be customized by the implementer, for example: an image block of a preset size around each connection point in the connection test can be set as a key area; or the area of ​​the minimum bounding box of every two connection points can be set as a key area. The areas at each scale of the clock drawing test and the pointer area in the center are set as key areas respectively; or the area of ​​the minimum bounding box of the clock is used as a key area. For different test tasks, the automatic feature extraction module extracts features from the image data by means of convolution, pooling and / or activation functions. The feature fusion module can fuse by splicing or channel superposition to obtain comprehensive features. The first evaluation module can evaluate the scores of multiple cognitive domains according to the comprehensive features through one or more fully connected layers. The second evaluation module can also evaluate the comprehensive score of cognitive ability according to the scores of multiple cognitive domains through one or more fully connected layers.

[0059] According to another embodiment of the present invention, the cognitive assessment model includes: a designated area feature extraction module, which is used to extract the first feature of the designated area set from the image data of the test feedback data using a neural network; a designated area identification module, which is used to determine the identification result of the designated area according to the first feature of each designated area; an automatic feature extraction module, which is used to automatically extract the second feature from the image data and associated data of the test feedback data using a neural network without setting the designated area, and extract the causal relationship between each brain region from the brain sensor data, wherein the associated data includes the test time and the test accuracy, and the brain sensor data is collected from the patient to be tested during the test process; a feature fusion module, which is used to fuse the identification result, the second feature and the causal relationship of the designated area to obtain a comprehensive feature; a first evaluation module, which is used to evaluate the scores of the patient to be tested in a preset multiple cognitive domains according to the comprehensive feature; and a second evaluation module, which is used to determine the comprehensive score of cognitive ability according to the scores of the multiple cognitive domains. The automatic feature extraction module of this embodiment is different from the previous embodiment, and additionally adds the function of extracting the causal relationship between each brain region from the brain sensor data, so that the cognitive assessment model can better evaluate the cognitive ability of the patient to be tested according to the causal relationship between each brain region.

[0060] In addition, the above embodiments may also be improved to obtain other implementations, for example: the associated data includes the test time but does not include the accuracy, and the model does not directly extract information from the accuracy but relies on the features extracted by the neural network for evaluation, further reducing the difficulty of implementation.

[0061] According to one embodiment of the present invention, before training the model, some normal people and patients with different degrees of cognitive impairment and patients diagnosed with Alzheimer's disease can be organized as sample personnel to participate in the production of data sets. Among them, each sample person completes various tests according to the test tasks selected by the doctor, and the doctor scores each sample person according to the test feedback data. The scoring results include the comprehensive score reference value of the sample person's cognitive ability and the score reference values ​​of multiple cognitive domains, and the scoring results are used as labels; and the identity information of the sample person and the brain sensor data during the test feedback are collected, and the sample person data is produced according to the identity information of the sample person, the test feedback data and the brain sensor data. Thus, a sample can be obtained according to each sample person, and a data set can be formed after being collected. Of course, a sample person can also cooperate with the doctor to participate in sample collection multiple times at different times to enrich the data set. Subsequently, a subset can be extracted from the data set as a training set to train the model.

[0062] Preferably, the cognitive assessment model is trained in the following manner: a training set is obtained, which includes multiple samples, each sample includes sample person data and a label, the sample person data includes the sample person's test feedback data and brain sensor data, and the label includes the sample person's cognitive ability comprehensive score reference value and the score reference values ​​of multiple cognitive domains; using the training set and a preset loss function, the cognitive assessment model is trained to evaluate the sample person's cognitive ability comprehensive score and the scores of multiple cognitive domains, wherein minimizing the loss function value is used as the optimization goal, and the parameters of the cognitive assessment model are updated according to the evaluated cognitive ability comprehensive score and the scores of multiple cognitive domains and the sample person's cognitive ability comprehensive score reference value and the score reference value of multiple cognitive domains recorded in the corresponding label.

[0063] The following is a schematic overall operation flow chart, a test phase workflow chart, and a cognitive assessment phase workflow chart of the system of the present invention during application, so as to roughly illustrate the assessment process. It should be understood that the following flow charts are only for illustration, and those skilled in the art can simply replace the order of some steps to form other similar operation processes.

[0064] According to one embodiment of the present invention, an overall operation flow chart of using the system of the present invention to perform cognitive assessment is as follows: Figure 3 As shown, including:

[0065] 1) When using it for the first time, you need to register as a doctor user;

[0066] 2) Enter your mobile phone number and password to log in to the system;

[0067] 3) Add patient information and select the patient to be tested;

[0068] 4) Enter the test, view the task list, or manage the doctor's information;

[0069] 5) View the analysis report of the tested tasks;

[0070] 6) Whether to choose to continue the test;

[0071] 7) Do you want to reselect the patient to be tested? If yes, go to 3), otherwise go to 8);

[0072] 8) End the cognitive assessment.

[0073] During the testing phase, you can choose different forms of multiple tasks to test. The workflow diagram of the testing phase is as follows: Figure 4 As shown, including:

[0074] 1) Select the task type, such as BBT, TMT, clock drawing test, pentagon test, complex figure test;

[0075] 2) Select the task test format, such as touch operation or photo upload;

[0076] 3) Select the specific configuration of the task, such as in BBT, select exercise or test, select the subtask type, select whether to add additional tasks, and select left hand or right hand;

[0077] 4) Start the test and continue until the timer runs out or the test is completed;

[0078] 5) Exit the test prompt and end this type of task;

[0079] 6) Whether to select other tasks;

[0080] 7) End the test.

[0081] In the cognitive assessment stage, you can view a variety of cognitive assessment reports. The cognitive assessment stage workflow diagram is as follows: Figure 5 As shown, it mainly includes:

[0082] 1) View the task list, which shows the completion status of all tasks;

[0083] 2) Check the completed tasks. The names of the tasks that have been completed during the testing phase are marked in black in the list and the checkboxes can be checked;

[0084] 3) Upload and analyze: upload the test data files of the selected tasks to the backend and analyze them using the cognitive assessment model;

[0085] 4) View the comprehensive cognitive report and obtain a comprehensive cognitive assessment report based on the uploaded tasks;

[0086] 5) Check the evaluation results of each task. For uploaded tasks, you can check the single task evaluation report;

[0087] 6) Review the assessment reports for each cognitive domain, including flexibility, memory, attention, perception, and agility;

[0088] 7) whether to return a comprehensive cognitive report;

[0089] 8) Whether to select more completed tasks. The higher the degree of task completion, the more accurate the results of the comprehensive cognitive report;

[0090] 9) End report review.

[0091] The system of the present invention is described below with a specific example. The system interface is simple in layout, with clear operation buttons and complete user management functions. Patients can complete tasks through simple clicks, slides and other gestures, which reduces the complexity of operations and ensures that the system is easy to operate and use. After the evaluation is completed, the system will automatically generate a detailed visual report, and users can intuitively view the evaluation results. The report content includes comprehensive evaluation results, analysis of each cognitive domain and fine-grained task performance, which helps doctors fully understand the patient's cognitive status.

[0092] 1. Equipment and devices

[0093] This example uses but is not limited to the Android development language and Xiaomi Tablet 6. The screen size of Xiaomi Tablet 6 is 11 and the screen resolution is 1880×1800. The cScreen desktop icon is as follows Figure 5 As shown, the main interface is as follows Figure 6 shown.

[0094] 2. Doctor registration and login

[0095] Doctors can register a system account by entering their mobile phone number, setting a password and security code. Figure 7 As shown, Figure 7 The cognitive dysfunction early screening system (cScreen) in the present invention refers to the cognitive assessment system of the present invention. Figure 7 a is a schematic diagram of the interface without filling in the registration information, b is a schematic diagram of the interface with the registration information filled in; c is a schematic diagram of the prompt interface when the password entered by the user does not meet the requirements; d is a schematic diagram of the prompt interface when the password entered by the user twice is inconsistent.

[0096] Then, the doctor's basic information is required, such as Figure 8 As shown. Among them, Figure 8 The a in the figure is a schematic diagram of the initial interface after the doctor user successfully registers, prompting the user to successfully register and the user ID; Figure 8 The diagram b in the figure is a schematic diagram of an interface for instructing a doctor user to fill in basic information, requiring the user to fill in name and hospital name.

[0097] After that, enter your phone number and password to enter the system, which supports functions such as remembering passwords and changing passwords. Fig. 9 shown.

[0098] 3. Patient management

[0099] Patient management can add, modify and delete patients, and visualize basic patient information (ID number, name, date of birth, gender), such as Fig.10As shown in the ad in the figure. After the doctor successfully enters the system, he can view all the patient information to be tested under the doctor user; if you need to add a new patient, click the "Add Patient" button, and the ID number needs to be filled in correctly, otherwise it cannot be added; if you need to modify or delete the patient, click the "Edit" button, and all the patient information can be modified in the middle of the pop-up window. The "Delete This Patient" button in the upper right corner of the pop-up window can delete the patient; click the "To be Tested" button corresponding to the patient to be tested to enter the task test.

[0100] 4. Task test

[0101] The task test module provides a variety of test tasks, which are mainly divided into five categories: BBT test, TMT test, clock drawing test, pentagon test and complex graphics test, such as Fig.11 As shown in af in the figure. It can be connected to external devices such as electroencephalogram (EEG) and functional near-infrared spectroscopy (fNIRS) to synchronously collect EEG signals. The "Change Patient" button can return to the patient management interface to reselect the patient to be tested, such as Figure 6 As shown in the figure; In addition to BBT, other test tasks include not only the screen sliding test method, but also taking photos and uploading them, such as Fig.12 In addition, in order to reduce the test deviation caused by the patient's unfamiliarity with the operation process, a test and practice mode can be provided for selection, such as Fig.12 As shown in b, the practice mode does not record the test feedback results and is only for the patient to be tested to practice.

[0102] 5. Mission Report

[0103] After the task test is completed, click "Analysis Report" in the bottom navigation bar to display the completion status of all subtasks, such as Fig.13 As shown; click "Upload and Analyze" to jump to the cognitive assessment comprehensive report, and view the special cognitive domain reports of flexibility, memory, attention, perception and agility. You can also view a single task report by clicking "Evaluation Results of Each Task"; by clicking the return button in the upper left corner, the cognitive assessment comprehensive report can be returned to the task list, and the cognitive domain assessment reports and single task assessment reports can be returned to the cognitive assessment comprehensive report, as shown Fig.14 As shown. Among them, Fig.14a is a schematic diagram of the comprehensive report of cognitive assessment. In addition to the score, you can also set the grade or level. For example, the brain ability level can be set to low, medium and high; the level of the cognitive domain can be set to lower, medium (can also be divided into upper medium, medium, lower medium), good and excellent; b is the agility (i.e. agility) cognitive domain report, which can provide the definition of the current cognitive domain, provide the corresponding levels of the same age by age, and the correspondence between different scores and various levels; c is the selection interface of each task evaluation result. It should be noted that each test task can set different test details to form different subtasks, and can be given different names (such as cTMT-Task 1, Task 2, etc.) to distinguish them. The rest are similar and will not be repeated; d is a single task report (specifically BBT-Task 1 task report).

[0104] 6. Doctor management

[0105] The basic information of doctors includes the doctor's name, mobile phone number and hospital to which they belong. The mobile phone number cannot be modified. To change the password, you need to enter the new password twice, which must not be the same as the original password. If you have made changes to the initial information and want to save it, click the "OK" button to successfully modify it. If you have made changes to the initial information but want to restore it to its original state, click the "Reset button" to restore it. Fig.15 As shown in the ad.

[0106] In addition, in some cases, the doctor module may not be set up, that is, cognitive assessment services can be directly provided to a large number of users. Each user can register independently on the system, and the system will provide cognitive assessment services to the user, and after the user participates in the relevant test tasks, the cognitive assessment report will be fed back.

[0107] It should be noted that although the above describes the various steps in a specific order, it does not mean that the various steps must be executed in the above specific order. In fact, some of these steps can be executed concurrently or even in a different order as long as the required functions can be achieved.

[0108] The present invention may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present invention.

[0109] A computer-readable storage medium may be a tangible device that holds and stores instructions used by an instruction execution device. Computer-readable storage media may include, for example, but are not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a protruding structure in a groove on which instructions are stored, and any suitable combination thereof.

[0110] The embodiments of the present invention have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A cognitive assessment system, characterized in that: include: A patient management module for storing and managing patient identity information and test-related information; A task testing module, used to provide a variety of test tasks for evaluating cognitive ability to the patient to be tested through a touch device, and obtain test feedback data of the patient to be tested through a touch screen and / or a camera; The task reporting module is used to evaluate the cognitive ability of the patient to be tested according to the test feedback data of the patient to be tested through a preset cognitive assessment model based on a neural network to generate a cognitive assessment report for the patient to be tested.

2. The system according to claim 1, characterized in that The cognitive assessment report includes scores of multiple cognitive domains and a comprehensive score of cognitive ability, wherein the cognitive assessment model determines the scores of the multiple cognitive domains according to the test feedback data, and determines the comprehensive score of cognitive ability according to the scores of the multiple cognitive domains.

3. The system according to claim 2, characterized in that The scores of the plurality of cognitive domains include scores of flexibility, memory, attention, perception and agility, or a combination thereof.

4. The system according to claim 2, characterized in that The cognitive assessment model includes: A designated area feature extraction module, used to extract a first feature of a set designated area from the image data of the test feedback data using a neural network; A designated area recognition module, used to determine a recognition result of the designated area according to the first feature of each designated area; An automatic feature extraction module, used for automatically extracting a second feature from the image data and associated data of the test feedback data by using a neural network without setting a designated area, wherein the associated data includes the test time and / or the test accuracy; A feature fusion module is used to fuse the recognition result of the specified area with the second feature to obtain a comprehensive feature; A first evaluation module is used to evaluate the scores of the patient to be tested in multiple preset cognitive domains according to the comprehensive characteristics; The second evaluation module is used to determine a comprehensive cognitive ability score based on the scores of the multiple cognitive domains.

5. The system according to claim 2, characterized in that The task test module also obtains brain sensor data of the patient to be tested when performing test feedback. The cognitive assessment model includes: A designated area feature extraction module, used to extract a first feature of a set designated area from the image data of the test feedback data using a neural network; A designated area recognition module, used to determine a recognition result of the designated area according to the first feature of each designated area; An automatic feature extraction module, used for automatically extracting a second feature from the image data and associated data of the test feedback data using a neural network without setting a designated area, and extracting the causal relationship between various brain regions from the brain sensor data, wherein the associated data includes the test time and / or the test accuracy; The feature fusion module is used to fuse the recognition result, the second feature and the causal relationship of the specified area to obtain a comprehensive feature; A first evaluation module is used to evaluate the scores of the patient to be tested in multiple preset cognitive domains according to the comprehensive characteristics; The second evaluation module is used to determine a comprehensive cognitive ability score based on the scores of the multiple cognitive domains.

6. The system according to claim 4 or 5, characterized in that: The cognitive assessment model is trained in the following manner: A training set is obtained, which includes a plurality of samples, each of which includes sample person data and a label, the sample person data includes test feedback data and brain sensor data of the sample person, and the label includes a comprehensive score reference value of the sample person's cognitive ability and score reference values ​​of multiple cognitive domains; The cognitive assessment model is trained by using the training set and the preset loss function to evaluate the comprehensive cognitive ability scores and scores of multiple cognitive domains of the sample persons, wherein minimizing the loss function value is taken as the optimization goal, and the parameters of the cognitive assessment model are updated according to the evaluated comprehensive cognitive ability scores and scores of multiple cognitive domains and the reference values ​​of the comprehensive cognitive ability scores and scores of multiple cognitive domains of the sample persons recorded in the corresponding labels.

7. The system according to any one of claims 1 to 5, characterized in that: The multiple test tasks include the following test tasks or a combination of the following test tasks: The box and block test task is used to provide a box containing multiple blocks, the interior of the box is divided into at least two areas, and the patient is instructed to move the blocks from one area to another area within a specified time; Wiring test, which is used to provide multiple connection points and instruct the patient to connect the wires according to the preset requirements to connect the various connection points; The clock drawing test is used to instruct the patient to draw a clock face with scales and hourly time numbers, and to instruct the patient to draw the hour and minute hands corresponding to a specified time on the clock face; The pentagon drawing test is used to instruct the patient to draw two partially overlapping pentagons as required, and the overlapping parts form a preset shape; The complex figure test is used to provide a complex figure to be copied, instruct the patient to observe and copy the complex figure, and instruct the patient to redraw the complex figure by recall after various time periods.

8. The system according to any one of claims 1 to 5, characterized in that: The brain sensor data includes electroencephalogram (EEG) and / or functional near infrared spectroscopy (FNIRS) data of the brain.

9. The system according to any one of claims 1 to 5, characterized in that: The system also includes: a doctor management module for managing doctor registration information; wherein the system is configured as follows: It is up to the doctor to add patients and select the test tasks that the patients need to participate in.

10. A touch control device, in which the system according to any one of claims 1 to 9 is deployed, and the touch control device is a device with a touch control screen and a camera.

Images