A method and system for constructing a writing ability brain fingerprint based on a cross-modal brain functional connectivity map

Abstract

The present invention discloses a method and system for constructing a writing ability brain fingerprint based on a cross-modal brain function connection map, which collects writing behavior data of users; collects neuroimaging data of users in writing task state and resting state; performs correlation analysis and identifies significantly correlated brain function connection features and behavior indicators; constructs cross-modal general brain function connection key features using brain function connection features and behavior indicators, and establishes a brain fingerprint prediction model; collects writing behavior data and neuroimaging data of new users, and verifies and optimizes the brain fingerprint prediction model; the system includes a behavior data collection device and a magnetic resonance imaging scanner, the behavior data collection device collects writing behavior data, and the magnetic resonance imaging scanner collects neuroimaging data of users in task state and resting state; establishes a brain fingerprint prediction model in a data processing device, and can accurately map out brain function connection features associated with individual differences in writing ability.

Description

Technical Field

[0001] The present invention belongs to the technical field of cognitive function training, and specifically relates to a method for constructing a writing ability brain fingerprint based on a cross-modal brain function connection map. Background Art

[0002] Writing involves a series of complex language, cognitive and motor processing. As a complex cognitive process unique to humans, writing will form individual-specific writing characteristics after long-term training, and show significant individual differences. The writing process and handwriting characteristics are closely related to the individual's physical health, personality and psychological characteristics, and psychological disorders. For example, writing is directly related to learning, and writing disorders are an important manifestation of children with learning difficulties. In addition, changes in handwriting can indicate the onset and progression of neurological diseases (such as Parkinson's disease, Alzheimer's disease, etc.) and psychological disorders (depression, anxiety). Moreover, the characteristics of individual writing are related to personality traits and can be applied to talent assessment. However, at present, individual differences in writing ability are often at the descriptive level, lacking scientific and objective evaluation methods, especially the lack of objective characterization of individual differences, and it is impossible to apply writing ability tests to individual academic level evaluation, physical health and mental health warning and treatment.

[0003] Studies have found that there are huge individual differences in human brain function and structure. Making full use of the information contained in individual differences will help to more accurately understand the brain mechanism of cognitive processing. From the perspective of clinical application, constructing brain-behavior associations at the individual level is a key step in applying brain imaging technology to disease biomarkers. With the advancement of magnetic resonance imaging data acquisition and analysis methods, it has become possible to study the brain mechanism of individual differences. First, the efficiency of data acquisition has been greatly improved. For example, the emergence of rapid scanning methods allows us to collect a large amount of individual data in a short period of time, which improves the signal-to-noise ratio. Secondly, data analysis methods are constantly developing, and researchers have developed computational methods that can construct behavior-brain relationships at the individual level. Therefore, the brain neuroscience of individual differences has gradually become a new trend in cognitive neuroscience, and has made great progress in different behaviors.

[0004] Among them, the prediction model based on the connection group is the most widely used and effective method, and has been successfully applied in predicting individual intelligence and attention. Previous studies have used brain function data of different modalities to construct brain-behavior relationships, but which modality of brain function can best represent individual differences in cognitive ability? It is still unclear. First of all, resting-state brain functional imaging does not require complex task design and is also suitable for different populations. These advantages make it the most commonly used individual difference brain indicator. Studies have found that resting-state brain function predicts attention, creativity, movement, emotion, working memory and other abilities at the individual level. However, resting-state functional data are obtained in an unconstrained state, and there are no clear requirements for brain input and output information. Therefore, it is difficult to distinguish between trait variables of interest and irrelevant state variables in resting-state brain function. Summary of the invention

[0005] In order to solve the above-mentioned technical problems existing in the prior art, the present invention provides a method for constructing a brain fingerprint of writing ability based on a cross-modal brain functional connection map, by combining behavioral tests and functional magnetic resonance imaging tests, using an individual prediction model based on the brain functional connection group, integrating behavioral data with functional magnetic resonance imaging data, and using correlation analysis to identify brain functional connection patterns that are closely related to writing ability, accurately mapping the brain functional connection characteristics associated with individual differences in writing ability.

[0006] The specific scheme adopted is as follows:

[0007] A method for constructing a writing ability brain fingerprint based on a cross-modal brain functional connectivity map, comprising:

[0008] Collecting user's writing behavior data on writing movement process and writing quality;

[0009] Collect neuroimaging data of users performing writing tasks and resting;

[0010] Perform correlation analysis on the collected writing behavior data and neuroimaging data to identify brain functional connectivity features and behavioral indicators that are significantly correlated with individual writing ability;

[0011] Using the identified brain functional connectivity features and behavioral indicators in writing task state and resting state, we constructed the key features of brain functional connectivity related to cross-modal general writing ability and established a brain fingerprint prediction model for predicting individual writing ability.

[0012] Collect writing behavior data and neuroimaging data of new users to verify and optimize the brain fingerprint prediction model.

[0013] Furthermore, the writing behavior data are data collected when performing direct copying tasks, delayed copying tasks, dictation test tasks and image copying test tasks; the writing movement process data are movement latency, writing movement time, stroke time interval, total time, writing speed, area and pen pressure collected when using a writing board and paper materials for testing; the writing quality is the handwriting evaluation writing quality based on paper version testing, including: stroke shape, connection fluency, size change, degree of inclination, spatial alignment, aesthetics, internal spacing of characters, stability and overall evaluation.

[0014] Preferably, the direct copying task requires the user to copy a series of specified Chinese characters. When executing the first 50% of the direct copying task, the font and handwriting of the user when copying are presented, and when executing the second 50% of the direct copying task, the font and handwriting of the user when copying are hidden.

[0015] Preferably, the delayed copying task is to display the Chinese characters and then cover them briefly, and then require the user to reproduce the Chinese characters from memory and write them.

[0016] Preferably, the dictation test task is to play the pronunciation of Chinese characters through a writing board and require the user to write them correctly on the writing board.

[0017] Furthermore, functional magnetic resonance imaging was used to collect neuroimaging data related to brain activity in various task states and resting states, and a whole-brain functional connectivity map was established using the selected whole-brain partition template as the region of interest. The corresponding whole-brain functional connectivity maps in the two modes of task state and resting state were intersected to form a cross-modal whole-brain functional connectivity map related to general writing ability.

[0018] Furthermore, combined with the established whole-brain functional connectivity map related to general writing ability, the functional connection strength of each brain region was calculated to construct the whole-brain functional connectivity; then, a correlation analysis was performed on the brain functional connectivity and writing behavior data to identify the brain functional connectivity characteristics and behavioral indicators that were significantly correlated with writing ability; using the brain functional connectivity characteristics and corresponding behavioral indicators, a brain fingerprint prediction model was constructed based on the connection group prediction model method.

[0019] On the other hand, the present invention also provides a writing ability brain fingerprint construction system based on a cross-modal brain functional connection map, which is connected to a data processing device, and the system includes:

[0020] A behavior data collection device, which collects the user's writing behavior data on the writing movement process and writing quality, and transmits it to the data processing device;

[0021] A magnetic resonance imaging scanner, connected to the data processing device, for collecting neurological imaging data of the user in a writing task state and a resting state;

[0022] The data processing device is provided with a data acquisition module, a writing ability assessment module, a multimodal magnetic resonance imaging data analysis module, a recognition module and a prediction model construction module;

[0023] The data acquisition module is used to collect the writing behavior data and neuroimaging data;

[0024] The writing ability assessment module performs a multi-dimensional assessment of writing ability based on the writing behavior data to obtain a writing behavior index including a writing movement process and a writing quality assessment;

[0025] The multimodal magnetic resonance imaging data analysis module obtains a cross-modal whole-brain functional connection map related to general writing ability based on the neuroimaging data in the writing task state and the resting state;

[0026] An identification module that combines writing behavior data and neuroimaging data to perform correlation analysis and identify brain functional connectivity features and behavioral indicators that are significantly associated with an individual's writing ability;

[0027] The prediction model building module uses the identified brain functional connectivity characteristics and behavioral indicators in the task state and resting state to construct key features of brain functional connectivity related to cross-modal general writing ability, and establishes a brain fingerprint prediction model for predicting individual writing ability.

[0028] Preferably, the behavior data collection device comprises:

[0029] The stylus has a pen core, a pressure sensor, an electromagnetic sensor and a handwriting ability measurement module, and is connected to a data processing device to obtain the writing behavior data of the user when writing;

[0030] A handwriting board, connected to the data processing device, for presenting the writing task during the user test;

[0031] a paper board, through which the user's handwriting is tested to evaluate the writing quality;

[0032] The reflector glasses are arranged on the magnetic resonance imaging scanner and are located in the head area of ​​the user when the user lies flat, and are used to reflect the image on the handwriting board into the user's field of vision.

[0033] Preferably, the handwriting board stores a direct transcription task module, a delayed transcription task module, a dictation test task module and a graphic transcription task module.

[0034] The technical solution of the present invention has the following advantages:

[0035] A. The brain fingerprint prediction model established by the present invention has obvious advantages over other traditional methods for evaluating writing ability. The present invention adopts a multimodal magnetic resonance method to collect the functional connection characteristics of the brain under writing tasks, and combines it with the corresponding writing behavior indicators. It can deeply analyze the brain activity patterns and deeply reveal the potential writing ability problems of users who may not show significant differences in the early stages. It can help us identify individuals who need attention before their behavior shows obvious differences.

[0036] B. The present invention, by using a writing board combined with magnetic resonance technology, can accurately capture differences in brain activity that are difficult to detect with traditional assessment methods. This not only provides an early warning for challenges that users may encounter in future learning processes, but also has a profound impact on their long-term academic and social development. In particular, early identification and timely intervention are crucial during the critical period of user growth, namely the sensitive period of writing skills.

[0037] C. The present invention uses brain imaging technology and combines behavioral performance with brain activity indicators to construct a brain fingerprint prediction model, which provides a new path for the evaluation of personalized writing ability. It proposes a new method for establishing a user's writing "brain fingerprint" in combination with brain function characteristics in different states, thereby providing a clear method path for personalized writing ability evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the specific embodiments of the present invention, the drawings required for use in the specific embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 A block diagram of the method for constructing a brain fingerprint of writing ability provided by the present invention;

[0040] Figure 2 A schematic diagram of the construction system provided by the present invention;

[0041] Figure 3 A schematic diagram of the process structure provided by the present invention;

[0042] Figure 4 A schematic diagram of the layout of the test content presented on the writing board;

[0043] Figure 5 A schematic diagram of the interface during the direct copying task test presented on the writing board;

[0044] Figure 6 A schematic diagram of the interface during the test of the graphic copying task presented on the writing board;

[0045] Figure 7 A schematic diagram of the equipment layout for writing tasks in a magnetic resonance imaging environment;

[0046] Figure 8 A schematic diagram of brain functional connections during writing;

[0047] Fig. 9 This is a schematic diagram of brain functional connectivity in a resting state;

[0048] Fig.10 A schematic diagram of brain functional connectivity in the comprehensive writing task state and resting state;

[0049] Fig.11 It is a "brain fingerprint" of individual differences in writing ability based on brain functional connections. DETAILED DESCRIPTION

[0050] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only 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.

[0051] like Figure 1 and Figure 3 As shown, the present invention provides a method for constructing a writing ability brain fingerprint based on a cross-modal brain functional connection map, comprising the following steps:

[0052]

S01

[0053] A comprehensive test of writing board and paper materials is used to evaluate the user's writing ability in multiple dimensions, including direct copying, delayed copying, dictation test, graphic copying and other test tasks, and to comprehensively collect multi-dimensional behavioral data on writing movement process and writing quality. The writing movement process evaluation indicators cover multiple dimensions, including movement latency, writing movement time, stroke time interval, total time, writing speed, area and pen pressure. At the same time, the handwriting quality is evaluated based on the paper version test, and the evaluation takes into account multiple dimensions such as stroke shape, connection fluency, size change, degree of inclination, spatial alignment, aesthetics, internal spacing of characters, stability and overall evaluation.

[0054] like Figure 4 As shown, the preferred writing behavior test of the present invention covers four writing tasks: direct copying, delayed copying, dictation test and graphic copying test, so as to comprehensively and deeply obtain the writing ability of the user.

[0055] The direct copying task requires the user to copy a series of specified Chinese characters. When executing the first 50% of the direct copying task, the handwriting of the user when copying is displayed on the handwriting board (preferably a tablet computer). When executing the second 50% of the direct copying task, the handwriting of the user when copying is hidden on the handwriting board. Figure 5 shown.

[0056] In the delayed transcription task, the user was shown a Chinese character on the handwriting board and then briefly covered, and then was asked to reproduce the character from memory.

[0057] In the dictation test task, the pronunciation of Chinese characters is played through the writing board, and the user writes it down correctly based on what he hears.

[0058] In the graphic copying task, users are required to copy a series of complex graphics, such as Figure 6 shown.

[0059] Through the above four writing tasks, the user's writing ability is deeply obtained. These test tasks not only examine the central process in the writing process, such as the extraction of glyph information and word selection, but also involve peripheral processes, namely the motor encoding and execution of glyph information. The direct transcription task examines the user's writing automation level when the writing result is invisible; the delayed transcription task evaluates the user's ability to extract and remember glyph information after the visual information disappears; the dictation test reveals the user's ability to convert auditory information into written output; the graphic transcription task evaluates the user's hand-eye coordination and fine motor ability in the process of transcribing complex visual patterns. The writing ability is comprehensively evaluated through writing movement indicators and writing quality.

[0060]

S02

[0061] like Figure 8 , Fig. 9 and Fig.10 As shown, the present invention uses functional magnetic resonance imaging (fMRI) technology to collect multimodal neuroimaging data when the user is performing a writing task and in a resting state, including the different task state data in the above step [S01]. These data help capture the brain activity pattern during the writing process, and also include the intrinsic functional activity pattern in the resting state.

[0062]

S03

[0063] Functional magnetic resonance imaging was used to collect neuroimaging data related to brain activity in various task states and resting states, and the selected whole-brain partition template was used as the region of interest (ROI) to construct the whole-brain functional connection spectrum in the writing task state and the resting state. Then, the correlation analysis of brain functional connection and writing behavior data was performed to identify brain functional connection features that were significantly related to writing ability, and the whole-brain functional connection maps related to behavior in the two modes of task state and rest state were intersected to form a cross-modal general writing ability-related whole-brain functional connection map, such as Fig.10 As shown. Of course, during the writing task, the user is responsible for collecting their brain activity data, while collecting the individual's intrinsic spontaneous brain activity data in a resting state, and performing necessary data preprocessing, including head motion correction, spatial standardization and smoothing, to ensure the quality of the data and the accuracy of the analysis.

[0064]

S04

S03

[0065] The present invention conducts a series of comprehensive assessments on the user's writing ability, and combines the brain activity data collected by the multimodal magnetic resonance imaging data analysis module to construct a writing "brain fingerprint" using the CPM connection model method. Fig.11 As shown, a new method has been developed to significantly improve the accuracy of individual differences in writing ability, thereby providing a clear methodological path for personalized writing ability assessment and application, with a prediction accuracy R value of 0.72.

[0066]

S05

[0067] The present invention aims to propose a new way to conduct writing assessment, which integrates multiple links such as data acquisition, behavioral data analysis, and the construction of individual writing brain functional connection maps. Specifically, through behavioral tests and functional magnetic resonance imaging scanning tasks, the behavioral data collection device uses handwriting board technology and paper board tests to collect and analyze data, focusing on the acquisition and analysis of movement parameters and the evaluation of writing results, so as to comprehensively evaluate the individual's writing skills. The functional magnetic resonance imaging task mainly records the brain area activation and brain functional connection construction during writing. The constructed brain fingerprint prediction model is aimed at the personalized brain functional connection map of Chinese character writing, with the aim of creating a "brain fingerprint" of the user's writing ability.

[0068] like Figure 2 As shown, the present invention also provides a writing ability brain fingerprint construction system based on a cross-modal brain functional connection map, which is connected to a data processing device, and the system includes: a behavioral data collection device and a magnetic resonance imaging scanner, the behavioral data collection device collects the user's writing behavior data on the writing movement process and writing quality, and transmits it to the data processing device; the magnetic resonance imaging scanner is connected to the data processing device, and is used to collect the user's neural imaging data in the writing task state and the resting state; the data processing device is provided with a data collection module, a writing ability evaluation module, a multimodal magnetic resonance imaging data analysis module, a recognition module and a prediction model construction module; the data collection module is used to collect writing behavior data and neural imaging data; the writing ability evaluation module performs a multi-dimensional evaluation of writing ability according to the writing behavior data, and obtains writing behavior indicators including writing movement process and writing quality evaluation; the multimodal magnetic resonance imaging data analysis module obtains a cross-modal writing ability-related general writing ability-related whole-brain functional connection map according to the neural imaging data in the two modes of writing task state and resting state, such as Fig.10 As shown in the figure, the common (overlapping) brain connections in the two modalities are used as the brain functional connection features of the model, which have higher accuracy in predicting behavior and can better represent the brain fingerprint of cognitive ability; the recognition module combines the writing behavior data and the neuroimaging data to perform correlation analysis, and identifies the brain functional connection features and behavioral indicators that are significantly correlated with the individual's writing ability; the prediction model construction module uses the identified brain functional connection features and behavioral indicators in the task state and the resting state to construct the key features of brain functional connections related to general writing ability across modalities, and establish a brain fingerprint prediction model for predicting individual writing ability.

[0069] The behavioral data collection device used in the present invention includes: a stylus, a handwriting board, a paper board and reflector glasses, wherein the handwriting board is preferably a tablet computer, and the paper board is a grid notebook. The stylus is composited with a refill, a pressure sensor, an electromagnetic sensor and a handwriting ability measurement module, and is connected to a data processing device to obtain the writing behavior data of the user when writing; the handwriting board is connected to the data processing device to present the writing task during the user test, and the handwriting board here can be integrated with the data processing device; the user's handwriting is tested on a paper board to evaluate the writing quality; in order to make it easier for the user to perform tasks and to complete the writing task comfortably while receiving a magnetic resonance imaging scan, the present invention uses reflector glasses, which are set on the magnetic resonance imaging scanner and located in the head area when the user is lying flat, to reflect the handwriting board image into the user's field of view, such as Figure 7 shown.

[0070] The user lies in the MRI scanner and keeps the body still to avoid any possible motion interference that may affect the data quality. In front of the user's line of sight, a specially designed handwriting board is placed on a specially designed desktop with adjustable height. In order to enable the user to clearly see the writing on the handwriting board while keeping the head in a fixed position, a specially designed pair of reflector glasses is used. This pair of glasses can reflect the image of the handwriting board on the desktop in front of the user into the user's field of vision, allowing the user to directly watch his or her writing movements without moving the head. This design not only improves the accuracy of writing data collection, but also greatly increases the user's comfort when performing writing tasks.

[0071] A direct transcription task module, a delayed transcription task module, a dictation test task module and a graphic transcription task module are stored in the handwriting board.

[0072] Preferably, the direct copying task module presents 24 low-frequency characters to the user in sequence, and the user needs to copy these low-frequency characters in the grid below. During this process, the user can see his or her handwriting in 12 tasks, but cannot see it in the other 12 tasks, so as to mainly test the user's ability to automate the writing process.

[0073] Preferably, the delayed copying task module presents 32 high-frequency characters to the user in sequence, and then these characters disappear, leaving only the grid. The user needs to write out the previously presented characters in the grid, and this step mainly tests the user's ability to extract Chinese characters.

[0074] Preferably, the dictation test task module first provides the user with an auditory voice of a word, then presents a word containing the word, and again provides the voice of the word. Finally, after a grid appears, the user needs to write the word in the grid, and this process mainly tests the user's ability to convert form and sound.

[0075] Preferably, the graphic copying task module presents 24 graphics of different complexity to the user, requiring the user to copy these graphics as accurately as possible. This step mainly tests the user's writing-related fine motor ability.

[0076] Preferably, the magnetic resonance task presents 24 characters to the user, 12 of which disappear immediately after presentation, and the user needs to write these characters in a grid, and the 12 characters do not disappear after presentation. The user copies the characters in the grid, and the user's brain activation data is recorded at the same time. Through this task, the user's brain area activation characteristics during the writing process are mainly examined, and a whole-brain functional connection map of the writing task state is constructed based on these characteristics. At the same time, in a resting state where the user is not doing any task, lie quietly for 8 minutes, collect the brain activity characteristics of the user's spontaneous activities, and construct a whole-brain functional connection map of intrinsic activities.

[0077] The writing process test uses a stylus and a tablet computer, where the tablet has a resolution of up to 2560x1600, a high refresh rate of 120Hz, a brightness of 500nit, and supports P3 color gamut, HDR10, and Dolby Vision, which greatly improves the accuracy of writing data recording. The stylus has a sampling rate of 240Hz and 4096 levels of pressure sensitivity, which ensures high-quality recording of lossless original writing data, facilitating detailed analysis and application of handwriting later.

[0078] The internal structure of the stylus is a composite of a refill, a pressure sensor, and an electromagnetic sensor. When the user applies pressure to write, the pressure sensor will deform, giving the sensor a certain range of motion to accurately capture the writing dynamics. In addition, the handwriting ability measurement covers an innovative and comprehensive handwritten Chinese character measurement technology, which converts the received magnetic induction signal into a data signal through a magnetic field signal receiving board. The electromagnetic signal receiving board is equipped with multiple sets of mutually perpendicular antenna arrays, which can trigger electromagnetic resonance when the stylus approaches and emits radio waves of a specific frequency, thereby capturing the feedback signal of the stylus. Through continuous signal sending and capturing cycles, the XY coordinate values ​​of the stylus relative to the receiving board are accurately calculated.

[0079] The present invention's assessment of the writing movement process covers latency, writing movement time, time interval, total time, writing speed, area and pen pressure:

[0080] Latency refers to the time interval from deciding to start writing to actually starting to write, which reflects the time an individual spends in the preparation stage before starting the writing task. This indicator is important for understanding an individual's reaction speed and preparation process. A shorter latency is generally considered a sign of efficient cognitive processing and good task preparation, which reveals an individual's reaction agility and preprocessing ability when facing a writing task.

[0081] Writing movement time represents the actual time required to complete the writing of each word without considering latency and interval time. This indicator helps to assess the fluency and speed of an individual's writing and is a key factor in measuring writing skills. From a psychological perspective, a shortened writing movement time may indicate that an individual has stronger motor control ability and a higher level of automated writing skills, reflecting their efficient psychomotor coordination and fine motor skills.

[0082] The writing interval refers to the pause time between each stroke during the writing process. This is crucial for analyzing the writing rhythm and the coherence of the action. Shorter intervals may mean that individuals have better action planning and sequence processing abilities when thinking and performing writing tasks, reflecting the effectiveness of higher-level cognitive processes and smooth thinking in writing tasks.

[0083] The total writing time is the total time required to complete a writing task, which directly reflects the writing efficiency. This indicator helps us comprehensively evaluate the overall time consumption of writing tasks, covering the entire cognitive and motor process from preprocessing to action execution. It reflects the individual's comprehensive ability in cognitive planning, motor preparation, execution and monitoring of writing tasks.

[0084] Writing speed, measured by the ratio of stroke length to the time required to write, is a key factor in assessing writing efficiency and fluency. In cognitive psychology, high writing speed is considered to be the result of the synergy of high-level cognitive skills and fine motor skills, including efficient integration of memory retrieval, visual-spatial processing, and hand motor control.

[0085] Area refers to the space covered by writing each character, which is of great significance for analyzing the spatial distribution and control ability of writing. In cognitive psychology, the size and consistency of the area occupied by writing can reflect the individual's visual spatial ability and the stability of writing movements.

[0086] Pen pressure refers to the force applied to the writing tool during writing. This indicator reveals an individual's writing habits and hand muscle control ability, and is also related to emotional state and stress level. Changes in pen pressure may be related to an individual's emotional fluctuations, tension, and cognitive burden, providing a window for understanding the relationship between an individual's psychological state and writing behavior.

[0087] At the same time, the assessment of writing quality includes stroke shape, smooth connection, size change, degree of inclination, spatial alignment, aesthetics, internal spacing of characters, stability, overall evaluation and total score of writing quality.

[0088] The stroke shape is related to the completeness and correctness of the strokes, reflecting the writer's mastery of the character structure and the accuracy of execution. The higher the score, the more complete the strokes are without missing parts and the fewer errors, showing the writer's deep understanding and high control ability of the Chinese character structure.

[0089] Connection fluency involves the naturalness and smoothness of the strokes and the connections between them. A high score indicates that the strokes connect naturally and smoothly during writing, without unnecessary pauses or stiffness, reflecting the writer's excellent movement coordination and coherent thinking.

[0090] Font size evaluates the appropriateness of the font, including too large or too small. Appropriate font size not only contributes to aesthetics, but also reflects the writer's good perception of space utilization and visual effects, as well as compliance with writing standards.

[0091] Font size variation focuses on the consistency of font size and evaluates whether there is an obvious trend of change. A high score for size consistency indicates that a stable font size can be maintained during writing, which shows the writer's control ability and consistency cognition during the writing process.

[0092] Spatial alignment examines the rationality of the spatial layout and combination of radicals. A high score indicates a reasonable layout, symmetry and harmony, demonstrating the writer's ability to organize space and his understanding of the beauty of Chinese character structure.

[0093] Internal spacing of characters refers to the appropriate spacing between strokes and radicals. Too large or too small spacing will affect the overall beauty and readability of the character. A high score for appropriate spacing reflects the writer's attention to details and adjustment ability.

[0094] Stability is judged from the thickness and strength of the handwriting. Handwriting with high stability indicates that the writer can maintain consistent strength and brushstrokes during the writing process, reflecting his or her good motor control and emotional stability.

[0095] The aesthetics measure the neatness and visual beauty of the font, including whether there are many erasures. A high score means that the font is neat and beautiful, with almost no erasures, which shows the writer's aesthetic ability and pursuit of writing quality.

[0096] The overall evaluation comprehensively considers the overall writing quality of the handwriting, reflecting the observer's subjective satisfaction with the handwriting. A high score represents excellent overall handwriting quality, reflecting the writer's ability to comprehensively use various writing skills and cognitive resources.

[0097] In the MRI writing task provided by the present invention, 24 Chinese characters are displayed to the user, divided into two groups: the first group includes 12 Chinese characters, which are displayed on the screen for a period of time and then disappear, requiring the user to rely on memory to accurately write these Chinese characters in the grid; the second group also includes 12 Chinese characters, but these Chinese characters remain displayed during the writing process, and the user needs to write these Chinese characters directly in the grid. During the entire writing process, the user's brain activity data is recorded.

[0098] The present invention uses the extracted brain functional connectivity features and behavioral indicators, and uses the `LinearRegression` module in Scikit-learn to build a multivariate linear regression model. In the model training process, cross-validation techniques such as k-fold cross-validation are used to ensure the stability and generalization ability of the model. By continuously adjusting and optimizing the model parameters, a connectome-based predictive modeling method (connectome-based predictive modeling, CPM) that can accurately predict the writing ability of an individual is successfully trained to build a brain fingerprint prediction model, providing an individual with a "brain fingerprint" of writing ability.

[0099] The matters not described in the present invention are applicable to the prior art.

[0100] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. However, the obvious changes or modifications derived therefrom are still within the scope of protection of the present invention.

Claims

1. A method for constructing a writing ability brain fingerprint based on a cross-modal brain functional connectivity map, characterized in that: include: Collecting writing behavior data about the writing movement process and writing quality of the user, wherein the writing behavior data is data collected when performing a direct copying task, a delayed copying task, a dictation test task, and an image copying test task; Collecting neuroimaging data of users in writing task states and resting states, specifically, using functional magnetic resonance imaging to collect neuroimaging data related to brain activity in various writing task states and resting states, and using the selected whole-brain partition template as the region of interest to establish a whole-brain functional connection map, and taking the intersection of the corresponding whole-brain functional connection maps in the task state and the resting state to form a cross-modal whole-brain functional connection map related to general writing ability; Perform correlation analysis on the collected writing behavior data and neuroimaging data to identify brain functional connectivity features and behavioral indicators that are significantly correlated with individual writing ability; Using the identified brain functional connectivity features and behavioral indicators in writing task state and resting state, we constructed the key features of brain functional connectivity related to cross-modal general writing ability and established a brain fingerprint prediction model for predicting individual writing ability. Collect writing behavior data and neuroimaging data from new users to verify and optimize the brain fingerprint prediction model; Combined with the established whole-brain functional connectivity map related to general writing ability, the functional connection strength of each brain region is calculated to construct the whole-brain functional connectivity; then, a correlation analysis is performed on the brain functional connectivity and writing behavior data to identify the brain functional connectivity characteristics and behavioral indicators that are significantly correlated with writing ability; using the brain functional connectivity characteristics and corresponding behavioral indicators, a brain fingerprint prediction model is constructed based on the connection group prediction model method.

2. The method for constructing a writing ability brain fingerprint according to claim 1, characterized in that: The writing movement process data refers to the movement latency, writing movement time, stroke time interval, total time, writing speed, area and pen pressure collected when testing with a writing board and paper materials; the writing quality refers to the handwriting evaluation writing quality based on the paper version test, including: stroke shape, connection fluency, size change, degree of inclination, spatial alignment, aesthetics, internal spacing of characters, stability and overall evaluation.

3. The method for constructing a writing ability brain fingerprint according to claim 2, characterized in that: The direct copying task requires the user to copy a series of specified Chinese characters. When executing the first 50% of the direct copying task, the font and handwriting of the user when copying are presented, and when executing the second 50% of the direct copying task, the font and handwriting of the user when copying are hidden.

4. The method for constructing a writing ability brain fingerprint according to claim 2, characterized in that: The delayed copying task is to display Chinese characters and then cover them briefly, and then require the user to reproduce the Chinese characters from memory and write them.

5. The method for constructing a writing ability brain fingerprint according to claim 2, characterized in that: The dictation test task is to play the pronunciation of Chinese characters through a writing board and require the user to write them correctly on the writing board.

6. A writing ability brain fingerprint construction system based on a cross-modal brain functional connectivity map, which is connected to a data processing device, characterized in that: The system comprises: A behavior data collection device, which collects the user's writing behavior data on the writing movement process and writing quality, and transmits it to the data processing device, wherein the writing behavior data is the data collected when performing the direct transcription task, the delayed transcription task, the dictation test task and the image transcription test task; A magnetic resonance imaging scanner, connected to the data processing device, for collecting neurological imaging data of the user in a writing task state and a resting state; The data processing device is provided with a data acquisition module, a writing ability assessment module, a multimodal magnetic resonance imaging data analysis module, a recognition module and a prediction model construction module; The data acquisition module is used to collect the writing behavior data and neuroimaging data; The writing ability assessment module performs a multi-dimensional assessment of writing ability based on the writing behavior data to obtain a writing behavior index including a writing movement process and a writing quality assessment; The multimodal magnetic resonance imaging data analysis module obtains a cross-modal whole-brain functional connection map related to general writing ability based on the neuroimaging data in the writing task state and the resting state; An identification module that combines writing behavior data and neuroimaging data to perform correlation analysis and identify brain functional connectivity features and behavioral indicators that are significantly associated with an individual's writing ability; A prediction model building module, which uses the identified brain functional connectivity features and behavioral indicators in the task state and resting state to construct the key features of brain functional connectivity related to cross-modal general writing ability and establish a brain fingerprint prediction model for predicting individual writing ability; Combined with the established whole-brain functional connectivity map related to general writing ability, the functional connection strength of each brain region is calculated to construct the whole-brain functional connectivity; then, a correlation analysis is performed on the brain functional connectivity and writing behavior data to identify the brain functional connectivity characteristics and behavioral indicators that are significantly correlated with writing ability; using the brain functional connectivity characteristics and corresponding behavioral indicators, a brain fingerprint prediction model is constructed based on the connection group prediction model method.

7. The writing ability brain fingerprint construction system according to claim 6 is characterized in that: The behavior data collection device comprises: The stylus has a pen core, a pressure sensor, an electromagnetic sensor and a handwriting ability measurement module, and is connected to a data processing device to obtain the writing behavior data of the user when writing; A handwriting board, connected to the data processing device, for presenting the writing task during the user test; a paper board, through which the user's handwriting is tested to evaluate the writing quality; The reflector glasses are arranged on the magnetic resonance imaging scanner and are located in the head area of ​​the user when the user lies flat, and are used to reflect the image on the handwriting board into the user's field of vision.

8. The writing ability brain fingerprint construction system according to claim 7, characterized in that: The handwriting board stores a direct transcription task module, a delayed transcription task module, a dictation test task module and a graphic transcription task module.

Images