Sustainable attention evaluation method based on near-infrared brain function imaging

Abstract

The invention discloses a continuous attention evaluation method based on near-infrared brain function imaging, relates to the technical field of attention evaluation, and particularly solves the problems that in existing continuous attention evaluation, only task performance of a current patient can be reflected, reasons causing abnormal task performance cannot be known, and in the prior art, the time or space resolution is low, and the accuracy is high. The invention provides a continuous attention evaluation method based on near-infrared brain function imaging in order to solve the problems that in the prior art, accurate position information of brain activities cannot be provided, and an unnatural body position needs to be kept for a long time, the invention provides a continuous attention evaluation method based on near-infrared brain function imaging, a near-infrared brain function imaging device is used for monitoring the blood oxygen level change condition of the related brain area of the subject during the task period; two aspects of data of behavioral indexes and iconography are combined together to assist in evaluating the persistent attention function of a subject, reasons causing behavioral performance are explained from the perspective of cerebral cortex functions, and a scientific basis is provided for formulating a personalized training scheme.

Description

Technical Field

[0001] The invention belongs to the technical field of attention evaluation, and in particular relates to a sustained attention evaluation method based on near-infrared brain function imaging. Background Art

[0002] Cognitive dysfunction refers to the abnormality in the process of the brain receiving, processing external information, and converting information into internal thinking, emotions and behavioral responses. It often occurs in patients with central nervous system damage, such as stroke, craniocerebral injury, spinal cord injury, etc. Sustained attention is the basis of many advanced cognitive functions. It refers to the ability to focus on task-related information for a long time. At present, the assessment of attention function mainly relies on scales and applications. These methods can only reflect the current task performance of patients, and cannot know the root cause of abnormal task performance. This is not conducive to observing the neural mechanism of improvement of attention function under rehabilitation treatment, especially in the cognitive function test software that has been published. Its program setting is only based on whether the subject makes a correct response after the stimulus is presented. There is a certain time delay between the brain processing the stimulus and generating the response. In the very short time after the stimulus appears, the response made by the subject is not the processing of the current stimulus, but may be caused by some internal or external factors, such as the delay in processing the previous stimulus.

[0003] At present, some scholars use brain functional imaging technology to reveal the state of the brain and its dynamic changes when cognitive activities occur. Therefore, electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) are the most common mainstream technologies at present, but they both have obvious shortcomings. The spatial resolution of EEG technology is low and cannot provide accurate location information of brain activity; the temporal resolution of fMRI technology is low, and the test requires the subject to lie in the magnetic resonance chamber. This forced unnatural posture will affect the brain activity of the subject when performing the task. On the other hand, both of these mainstream technologies require strict restrictions on the shaking of the subject's head and limbs, and are not suitable for monitoring the task status for a long time;

[0004] Therefore, it is urgent to design a sustained attention evaluation method based on near-infrared brain functional imaging to solve the above problems. Summary of the invention

[0005] The above-mentioned existing technologies can only reflect the current patient's task performance, but cannot reveal the root cause of the abnormal task performance. In addition, the current brain functional imaging technology has low spatial resolution and low temporal resolution when used, and requires a long time to ensure that the subject's head and limbs cannot shake.

[0006] In order to solve the above problems, the technical solution adopted by the present invention is as follows:

[0007] A sustained attention evaluation method based on near-infrared brain functional imaging, characterized in that:

[0008] S1. Determine the brain function task paradigm and set the task specific parameters;

[0009] S2. Determine the anatomical partition of the subject's cerebral cortex corresponding to the test channel, and use a near-infrared brain functional imaging device to monitor the brain function state of the subject during the task of step S1, and obtain near-infrared brain functional imaging data and behavioral data during the task;

[0010] S3. judging the near-infrared brain function imaging data to obtain reliable near-infrared brain function imaging data;

[0011] S4. Preprocess the reliable near-infrared brain functional imaging data;

[0012] S5. Based on the processed near-infrared brain function imaging data, the near-infrared brain function imaging data of the same category of test channels are superimposed and averaged to calculate the corresponding integral value and center of gravity value;

[0013] S6. Using the behavioral data in step S2 and the integral value and center of gravity value in step S5 as indicators, evaluate the subject's sustained attention function status.

[0014] Preferably, in step S1, the brain function task paradigm is a digit flash task.

[0015] Preferably, in step S2, the process of determining the anatomical partition of the subject's cerebral cortex corresponding to the test channel of the near-infrared brain function imaging device includes:

[0016] S201. Using the position coordinates of the skull reference point, establish a correspondence between the subject and the skull in a certain individual R1 structural head MRI image in the database, and based on the correspondence, map the positions of all the transmitters, receivers, and each test channel on the skull in the individual R1 head structural MRI image;

[0017] S202. Based on the individual R1 head structural MRI image and the virtually placed optode positions, estimate the corresponding cortical positions and MNI coordinates of all test channels in the individual R1 head structural MRI image using a balloon inflation algorithm;

[0018] S203. The remaining individuals R2-R in the database n After all steps S201 and S202 are repeated, each test channel obtains the same number of corresponding cortical positions and their MNI coordinates as the number of individuals in the database;

[0019] S204. Obtain a final MNI coordinate based on the average of several MNI coordinates, and determine the anatomical partition of the cerebral cortex corresponding to the test channel according to the final MNI coordinate.

[0020] Preferably, in step S2, the behavioral data include: number of positive responses, number of incorrect responses, CANCEL, accuracy rate, hit rate, average reaction time, standard deviation and coefficient of variation.

[0021] Preferably, in step S3, the process of obtaining credible near-infrared brain function imaging data includes discarding the near-infrared brain function imaging data corresponding to the behavioral data with a coefficient of variation greater than 15%.

[0022] Preferably, in step S4, the process of preprocessing the credible near-infrared brain functional imaging data includes:

[0023] S401. discarding the test channels in the reliable near-infrared brain functional imaging data with obvious motion artifacts and poor signal quality;

[0024] S402. Based on the modified Beer-Lambert law, the original light intensity data is converted into relative change data of cerebral blood oxygen concentration, and the relative change data of cerebral blood oxygen concentration includes: oxygenated hemoglobin concentration Δ[oxy-Hb] signal, deoxygenated hemoglobin concentration Δ[deoxy-Hb] signal and total oxygenated hemoglobin concentration Δ[total-Hb] signal;

[0025] S403. Use a 0.1 Hz low-pass filter to remove the noise components with obvious periodic fluctuation characteristics in the near-infrared brain functional imaging data;

[0026] S404. Use the Savitzky-Golay method to smooth the data once, and the number of smoothing points is set to 5;

[0027] S405. Take the average value of the relative change signal of the blood oxygen concentration within 15 seconds before the start of the task and perform baseline correction.

[0028] Preferably, in step S5, the process of calculating the integral value and the center of gravity value includes:

[0029] S501. Based on the pre-processed near-infrared brain functional imaging data, according to the probabilistic registration result of the positioning data, the test channels with good quality are divided into different categories according to their Brodmann partitions and functions, and then the data of the test channels of the same category are superimposed and averaged;

[0030] S502. Based on the superimposed and averaged data, calculate the corresponding integral value and center of gravity value.

[0031] Preferably, in step S502, the integral value calculation formula is:

[0032]

[0033] Where f(t) represents the signal function of the relative change of oxygenated hemoglobin concentration, t represents time, and t task start Represents the time when the task starts, t task end represents the time point when the task ends, and d represents the differential.

[0034] Preferably, in step S502, the center of gravity value calculation process includes:

[0035] S502-1. Calculate the total area of ​​the positive change region under the fNIRS positive signal during the entire test period;

[0036] S502-2. Set a threshold value, which is 1 / 2 of the total area of ​​the positive change region;

[0037] S502-3. From the start time of the test, gradually increase the time t, and calculate the area under the curve of the fNIRS positive signal up to the current time t. When the cumulative area is greater than or equal to the threshold, the time point is the centroid value, and its expression is:

[0038]

[0039] Where f(t) represents the signal function of the relative change of oxygenated hemoglobin concentration, and when f(t)>0, t represents time, t test start Represents the time point when the test starts, t test end Represents the time point when the test ends, t centroid value is the time point corresponding to the centroid value.

[0040] The beneficial effects of the present invention are as follows: the present invention discloses a method for evaluating sustained attention based on near-infrared brain function imaging. Compared with the prior art, the present invention has the following improvements:

[0041] (1) The present invention provides a simple and easy-to-operate sustained attention evaluation method. It uses behavioral indicators of a digital flash task, including accuracy, hit rate, average reaction time, etc., to evaluate the subject's ability to recognize target stimuli and sustained attention performance when receiving multiple stimuli presented continuously. At the same time, a near-infrared brain functional imaging device is used to monitor the changes in the blood oxygen level of the subject's relevant brain areas during the task to reflect their brain function status. By combining behavioral and imaging data, the subject's attention function is more comprehensively evaluated, and the reasons for the behavioral performance are further explained from the perspective of cerebral cortical function, providing a scientific basis for formulating personalized training programs.

[0042] (2) When using it, the patient is tested before and after treatment to evaluate the effect of rehabilitation training and explore its related mechanisms for improving attention function. For patients with severe cognitive dysfunction, short-term training may not be enough to significantly improve their task performance. Behavioral indicators have not changed significantly, but changes in brain function may have occurred. Changes in the intensity and speed of activation of task-related brain areas can be reflected by integral values ​​and center of gravity values. These indicators are of great significance for evaluating the efficacy. On the other hand, this method also has great potential in predicting the prognosis of cognitive function in patients with central nervous system damage. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a flow chart of the sustained attention evaluation method based on near-infrared brain functional imaging of the present invention;

[0044] Figure 2 The task paradigm of the digital flash task of the present invention;

[0045] Figure 3 This is a diagram of the photoelectrode arrangement of the present invention;

[0046] Figure 4 It is the Broadmann partition diagram corresponding to the test channel of the present invention;

[0047] Figure 5 This is a schematic diagram of superimposed averaging of similar channels of the present invention;

[0048] Figure 6 It is a schematic diagram of the integral value of the present invention;

[0049] Figure 7 This is a schematic diagram of the center of gravity value of the present invention;

[0050] Figure 8 This is a flow chart of the sustained attention evaluation method based on near-infrared brain functional imaging of the present invention. DETAILED DESCRIPTION

[0051] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0052] Functional near-infrared spectroscopy (fNIRS) is a non-invasive functional imaging technology. When performing a specific task, task-related neurons start to work, and the local metabolic activity of the brain tissue where the neurons are located increases. Since the energy required by neurons is almost entirely supplied by the aerobic metabolism of glucose in real time, metabolic activity will cause a decrease in the HbO2 concentration and an increase in the HbR concentration in the local tissue blood. These changes in blood oxygen concentration will cause local vasodilation, increase cerebral blood flow and cerebral blood volume, and increase the amount of oxygen provided by the increased cerebral blood flow far greater than the actual oxygen consumption of neuronal activity, ultimately leading to an increase in the local tissue HbO2 concentration. In other words, we can indirectly reflect neural activity through changes in local blood oxygen concentration in brain tissue. fNIRS detection is based on the physiological principle of neurovascular coupling mechanism and the physical principle of changes in the optical properties of brain tissue caused by changes in local blood oxygen concentration. The latter uses the modified Beer-Lambert law to establish the relationship between near-infrared light attenuation and changes in blood oxygen concentration. fNIRS has high temporal and spatial resolutions, and its most significant advantage is its high ecological validity, which allows subjects to be tested in a near-natural setting, and the results can more realistically reflect the brain's functional state during the task. In addition, fNIRS is much less sensitive to head movement than other brain functional imaging techniques, and subjects can blink, speak, and adjust their neck posture appropriately. Therefore, the comfort and acceptability of this technology make it more suitable for longer-term task monitoring.

[0053] Embodiment 1:

[0054] Refer to the attached Figure 1-8 A method for evaluating sustained attention based on near-infrared brain functional imaging is shown, comprising:

[0055] S1. Determine the brain function task paradigm and set the task specific parameters;

[0056] When using near-infrared brain function imaging technology to detect the brain function status of the subjects completing cognitive tasks, the subjects are often unable to maintain a relatively stable level of attention throughout the test due to fatigue or emotional state, and may experience distraction, random thoughts, and excessive tension. This can cause large differences in the changes in brain blood oxygen levels detected by near-infrared when the test is repeated multiple times, thereby affecting the credibility of the results. In order to ensure the accuracy and reliability of the test results, this embodiment uses the coefficient of variation of the digital flash task as an evaluation indicator to screen the subject's data and eliminate test errors caused by large fluctuations in attention levels. Therefore, in this step, the digital flash task is selected as the brain function task paradigm, and in this embodiment, the attached Figure 2 The task paradigm;

[0057] The specific content and process of the digital flash task test are as follows: the numbers 0-9 appear randomly on the display screen, and the subjects are asked to press the space bar as quickly as possible to respond when the target number "7" appears, and to inhibit the response when other numbers appear. The task duration and the ratio of target stimulation to non-target stimulation can be set according to the test needs;

[0058] The test results of the digital flash task include the number of positive responses, the number of incorrect responses, CANCEL, accuracy, hit rate, average reaction time, standard deviation and coefficient of variation. The number of positive responses refers to the number of correct responses, including the number of correct responses to target stimuli and the number of non-responses to non-target stimuli; the number of incorrect responses refers to the number of incorrect responses, including the number of missed target stimuli and the number of responses to non-target stimuli; CANCEL refers to the number of responses made within 110ms; the accuracy is calculated as the number of positive responses (to target stimuli) / the total number of (target stimulus presentations) × 100%. The accuracy is the proportion of subjects who correctly respond to target stimuli in the task, which mainly focuses on the subject's ability to recognize target stimuli, and does not involve responses to non-target stimuli; the hit rate is calculated as the number of positive responses (to target stimuli) / the total number of responses (i.e., the number of actual key presses, including the number of times the target stimuli are seen). The hit rate is the proportion of correct responses to stimuli in the task. As a comprehensive indicator, it not only takes into account the subject's ability to recognize target stimuli, but also includes the ability to distinguish non-target stimuli. The average reaction time is the average reaction time to the target stimulus (excluding CANCEL), reflecting the average reaction speed when making a correct response to the target stimulus. The standard deviation is the value indicating the degree of dispersion of the reaction time. The coefficient of variation is calculated as standard deviation / average reaction time×100%. When the coefficient of variation is greater than 15%, the data obtained are considered abnormal and should be discarded.

[0059] The target stimulus reaction time is the time for the subject to press the key as quickly as possible to react after seeing the target number "7" appear on the screen, that is, the time required for the subject to receive, process the stimulus and produce a reaction. It should be noted that when the reaction time of any stimulus is less than 110ms, the reaction is counted as CANCEL, and the reaction time is not included in the target stimulus reaction time, nor is it involved in the calculation of the average reaction time, standard deviation and coefficient of variation. The reason is that there is a certain time delay between the brain processing the stimulus and producing the reaction. In a very short time after the stimulus appears, the reaction made by the subject is not the processing of the current stimulus, but may be caused by some internal or external factors (such as the delay in the processing of the previous stimulus). In order to avoid the impact of premature reactions on the test results, the present invention sets a threshold value for the reaction time, and counts the number of times a reaction is made within 110ms as CANCEL, excluding those obviously too fast reactions, and more accurately assessing the true reaction speed and accuracy of the subject.

[0060] Specifically, in this embodiment, the duration of the digital flash task is set to 5 minutes, with a 30-second baseline period before and after the task, and a total time of 6 minutes. During the baseline period, the subjects were asked to sit quietly in front of the screen and stare at the white cross in the center of the screen. The application of a long (300-second) simple cognitive activation task can induce compensatory activation in PSCI patients with right hemisphere damage, and can also provide more complete and detailed time course characteristics of frontal lobe blood oxygen response.

[0061] The subjects who perform the digital flash task are patients with cognitive dysfunction, including those with stroke, brain trauma, spinal cord injury and other diseases. In addition to cognitive dysfunction, these patients often have motor dysfunction to varying degrees. Regarding the response to the target stimulus (i.e., the target number), hemiplegic patients use the healthy hand to complete the test; paraplegic patients can use their remaining motor functions (such as wrist and elbow flexion and extension) and use an external USB large button to simulate the space bar to complete the test; patients with severe motor dysfunction can respond verbally, and the examiner will respond instead of pressing the button.

[0062] S2. Determine the anatomical partition of the subject's cerebral cortex corresponding to the test channel, and use a near-infrared brain functional imaging device to monitor the brain function state of the subject during the task of step S1, and obtain near-infrared brain functional imaging data and behavioral data during the task;

[0063] Specifically, a 3D positioning device is used to record the three-dimensional position coordinates of all skull reference points, the transmitter and receiver, and the test channel, and the MNI coordinates corresponding to each test channel are obtained through the general registration method, and then the anatomical division of the cerebral cortex corresponding to the MNI coordinates is obtained; according to the international 10-20 reference system, the optode of the near-infrared brain functional imaging device is made to cover the prefrontal cortex of the subject's brain; the optodes are arranged on the optode cap in 2×8, such as Figure 3 As shown, it consists of 8 transmitting electrodes and 8 receiving electrodes, with a total of 22 channels, which monitor the brain function state of the subject during the task of step S1, and obtain near-infrared brain function imaging data and behavioral data during the task;

[0064] In this step, the near-infrared brain function imaging device uses the research near-infrared brain function imaging system Light-NIRS (Shimadzu Enterprise Management (China) Co., Ltd. / Shimadzu (Hong Kong) Co., Ltd.), which has three wavelengths of near-infrared light, namely 780nm, 805nm and 830nm. The sampling rate of the device is 13.33Hz, and the photoelectrodes are arranged in 2×8 on the photoelectrode cap, consisting of 8 emitters and 8 receivers, forming 22 channels; the spacing between the emitters and receivers is 3cm, and the detection depth is about 2-2.5cm below the cortex. According to the international 10-20 reference system, the bottom of the photoelectrode array is placed along the Fp1-Fp2 line, and Fpz is located at the midpoint of the bottom line to ensure that the photoelectrode covers the prefrontal cortex. The NirSpace 3D spatial positioning system is used to collect the three-dimensional position coordinates of the skull reference point, the emitter and receiver, and the test channel. The skull reference points specifically include: Nz, AL, AR, Iz and Cz points, among which Nz point is the depression at the root above the bridge of the nose; AL and AR points are the points in front of the left and right ears, that is, the most prominent parts of the left and right tragus; Iz point is the occipital protuberance; Cz point is the central point, which is the intersection of the two baselines Nz-Cz-Iz and AL-Cz-AR; the position of the test channel is the midpoint of the line connecting the paired transmitter and receiver. The MNI coordinates corresponding to each test channel are obtained by the general registration method, and the anatomical partition of the cerebral cortex corresponding to the test channel is obtained according to the MNI coordinates, so as to facilitate the superposition and averaging according to the anatomical partition and function of the cerebral cortex to which the test channel belongs, and calculate its corresponding integral value and center of gravity value;

[0065] The specific steps to determine the anatomical partition of the cerebral cortex corresponding to the device test channel are:

[0066] ① Using the position coordinates of the skull reference points, establish the correspondence between the subject and the skull in an individual R1 head structure MRI image in the database, and based on this correspondence, map all the emitters and receivers, as well as the positions of each measurement channel on the skull in the R1 individual head structure MRI image (it can be understood as placing the optical electrodes on the subject's head virtually on the R1 head). If it is necessary to further improve the accuracy of the correspondence between the subject's head and the skull in the individual structural MRI image, more reference points of the 10-20 reference system can be added on the basis of the 5 skull reference points; MRI images are magnetic resonance imaging;

[0067] ②Based on the skull and its virtually placed optode positions (emitter and receiver positions) in the individual R1 head structural MRI image, the balloon inflation algorithm was used to estimate the corresponding cortical positions and MNI coordinates of all test channels in the individual R1 head structural MRI image.

[0068] ③ The remaining individuals in the database (R2-R n) Repeat steps ① and ② in full, and each measurement channel will obtain the corresponding cortical position and its MNI coordinates that are the same as the number of individuals in the database. The average of several MNI coordinates is the most likely position of the test channel in the subject's cortex. Therefore, the average of the MNI coordinates is used as the final MNI coordinate corresponding to the test channel, and then the anatomical partition of the cerebral cortex corresponding to the test channel is determined according to the final MNI coordinate. The MNI coordinates corresponding to the test channel of this embodiment and its corresponding anatomical partition of the cerebral cortex (Broadmann partition) are shown in Tables 1 and Figure 4 As shown;

[0069] Table 1 MNI coordinates and Broadmann partitions corresponding to the test channels

[0070]

[0071]

[0072] S3. judging the near-infrared brain function imaging data to obtain reliable infrared brain function imaging data;

[0073] Specifically, the coefficient of variation is used to judge the credibility of near-infrared brain function imaging data, and reliable infrared brain function imaging data is obtained; the near-infrared brain function imaging data corresponding to the coefficient of variation greater than 15% is discarded, and the near-infrared brain function imaging data corresponding to the coefficient of variation less than 15% is further analyzed. This operation ensures that the obtained test data can reflect the relatively stable attention level of the subject, and avoids the subject's distraction, random thoughts or excessive tension during the task due to fatigue or emotional state, so that the data cannot truly reflect the subject's sustained attention level, thereby affecting the credibility of the results.

[0074] S4. Preprocess the reliable near-infrared brain functional imaging data; Figure 5 The following is a schematic diagram of removing two channels (ch4 and ch19) in Table 1 and classifying them; the specific steps of preprocessing are:

[0075] ①Discard the test channels in the reliable near-infrared brain functional imaging data with obvious motion artifacts and poor signal quality.

[0076] ② Based on the modified Beer-Lambert law, the original light intensity data collected by the near-infrared brain functional imaging device was converted into the relative change data of cerebral blood oxygen concentration. The relative change data of cerebral blood oxygen concentration included: oxygenated hemoglobin concentration Δ[oxy-Hb] signal, deoxygenated hemoglobin concentration Δ[deoxy-Hb] signal and total oxygenated hemoglobin concentration Δ[total-Hb] signal. In further analysis, we focused on the change of Δ[oxy-Hb] signal because the oxy-Hb signal is more sensitive to changes in cerebral blood flow and has a higher signal-to-noise ratio and retest reliability.

[0077] ③ Use a 0.1 Hz low-pass filter to remove noise components with obvious periodic fluctuation characteristics in near-infrared brain functional imaging data, including machine noise and physiological noise, such as heart rate (about 1 Hz), breathing (about 0.2-0.3 Hz), blood pressure fluctuation period Mayer wave (about 0.1 Hz), etc.

[0078] ④ Use the Savitzky-Golay method to smooth the data processed in step ③ once, and the number of smoothing points is set to 5.

[0079] ⑤ Take the average value of the relative change signal of blood oxygen concentration within a few seconds before the start of the task (0-5s, 0-10s or 0-15s can be selected) to perform baseline correction; the purpose of baseline correction is to remove the baseline signal from the original blood oxygen data. After correction, the baseline period data is around 0, and the change of blood oxygen level during the task relative to the baseline period can be seen; this embodiment selects the average value of the relative change signal of blood oxygen concentration within 15 seconds before the start of the task for baseline correction.

[0080] S5. Based on the processed near-infrared brain function imaging data, the near-infrared brain function imaging data of the same category of test channels are superimposed and averaged to calculate the corresponding integral value and center of gravity value;

[0081] According to the anatomical division and function of the cerebral cortex to which the test channel belongs, the near-infrared brain functional imaging data of the same category of test channels are superimposed and averaged, and the corresponding integral value and center of gravity value are calculated; specifically,

[0082] ① Based on the pre-processed near-infrared data, according to the probabilistic registration results of the positioning data in Table 1, the test channels with good quality are divided into different categories according to their Brodmann partitions and functions, and then the data of the channels of the same category are superimposed and averaged; the results of dividing different categories are as follows Figure 5 As shown; in the actual operation process, the left and right brain prefrontal lobe channels can also be superimposed and averaged respectively, where the right prefrontal lobe includes CH1, CH2, CH3, CH8, CH9, CH10, CH11, CH16, CH17, and CH18, and the left prefrontal lobe includes CH5, CH6, CH7, CH12, CH13, CH14, CH15, CH20, CH21, and CH22;

[0083] ② Based on the superimposed and averaged data, the integral value and center of gravity value are calculated to obtain the integral value and center of gravity value corresponding to each category of channels. Different categories of channels correspond to various functional areas of the brain;

[0084] Specifically, after channel superposition and averaging, the integral values ​​and center of gravity values ​​of the bilateral dorsolateral prefrontal cortex area (BA46, BA9), frontopolar area (BA10), motor language center (BA44, BA45), premotor area and supplementary motor area (BA6) can be obtained. Among them, the dorsolateral prefrontal cortex area plays a very important role in the brain's higher-level cognitive functions (such as working memory, attention and executive functions). The frontopolar area is the key activated brain area for prospective memory (a memory pointing to the future, that is, remembering to perform a specific activity at a specific moment or situation in the future). The main function of the motor language center is the expression of language. The premotor area and supplementary motor area are mainly responsible for regulating the related movements of the trunk muscles, planning the completion sequence of complex movements and coordinating the movements of both sides.

[0085] The integral value represents the sum of signal changes during the execution of the task, and a larger integral value means a greater activation of the brain area related to the task.

[0086] The integral value calculation formula is:

[0087]

[0088] Where f(t) represents the signal function of the relative change of oxygenated hemoglobin concentration, t represents time, and t task start Represents the time when the task starts, t task end represents the time point when the task ends, and d represents the differential;

[0089] It should be noted that when calculating the integral value, the baseline period needs to be excluded. For example, if the total task duration of the subject is 0-360s, and the baseline period is 30s before and after the task, then in the integral value calculation, the Δ[oxy-Hb] in the oxygenated hemoglobin concentration range of 30-330s is summed, which can also be understood as calculating the area under the curve of the continuously changing blood oxygen curve in the task execution interval [30,330]. Figure 6 The shaded area shown represents the integrated value, and the unit is (mM*cm)×s.

[0090] The center of gravity value is the time when the area under the fNIRS positive signal curve reaches half of the total area of ​​the positive change area during the entire test period (including the baseline period before and after). The smaller the center of gravity value, the faster the brain area responds to the task. The fNIRS positive signal refers to the signal with a Δ[oxy-Hb] (oxyhemoglobin concentration signal) value greater than 0;

[0091] The calculation steps are as follows: ① Calculate the total area of ​​the positive change area under the fNIRS positive signal during the entire test period. ② Set the threshold, which is 1 / 2 of the total area of ​​the positive change area. ③ From the start time of the test, gradually increase the time t, and calculate the area under the curve of the fNIRS positive signal up to the current time t. When the cumulative area is greater than or equal to the threshold, the time point is the centroid value. The expression is: When f(t)>0

[0092]

[0093] Where f(t) represents the signal function of the relative change of oxygenated hemoglobin concentration, t represents time, and t test start Represents the time point when the test starts, t test end Represents the time point when the test ends, t centroid value is the time point corresponding to the centroid value;

[0094] If the entire test duration of the subject is 360s, the time point represented by the centroid value divides the area of ​​the positive change region in the interval [0,360] equally, such as Figure 7 Point C shown represents the center of gravity value in s.

[0095] S6. A process of evaluating the subject's sustained attention function using the behavioral data in step S2 and the integral value and center of gravity value in step S5 as indicators;

[0096] The behavioral data of the subjects completing the step S1 task paradigm include: the number of positive responses, the number of incorrect responses, CANCEL, accuracy, hit rate, average reaction time, standard deviation and coefficient of variation; the integral value and center of gravity value are the imaging results; in this step, the behavioral data of the task paradigm and the corresponding integral value and center of gravity values ​​obtained from the multi-category channel data need to be compared. Through the comparison of data, the sustained attention function of the subjects can be objectively evaluated, which can be used in future clinical environments to assist in identifying sustained attention dysfunction caused by various diseases.

[0097] During the evaluation process, the subjects need to be tested multiple times at different time points, such as at admission, one week after treatment, and two months after treatment. The effect of rehabilitation training can be evaluated by comparing the test data before and after treatment. If the subject's accuracy and hit rate are improved and the average reaction time is shortened, it means that the treatment at this stage can improve their sustained attention function. Changes in the integral value and center of gravity value of each brain region can reveal the relevant mechanisms of changes in attention function. For patients with severe cognitive dysfunction, short-term training may not be enough to improve their task performance, and behavioral indicators have not yet shown obvious changes, but changes in brain function may have occurred. The integral value and center of gravity value of task-related brain areas can reflect changes in the intensity and speed of brain activation, which is of great significance for judging the therapeutic effect.

[0098] Embodiment 2:

[0099] Example 2 is to use the evaluation method of Example 1 to evaluate patients with post-stroke cognitive impairment (PSCI) due to right brain injury;

[0100] 1. Select subjects;

[0101] The present invention included 16 patients with post-stroke cognitive impairment (PSCI) due to right brain damage admitted to the Occupational Therapy Department of Beijing Boai Hospital from May 2020 to May 2022 to form the right hemisphere damage (RHD) group, and 32 age-matched healthy volunteers were selected as the control group.

[0102] Inclusion criteria were: ① meeting the diagnostic criteria for stroke and first stroke, unilateral cerebral hemisphere lesions; ② meeting the diagnostic criteria for PSCI; ③ 14 to 150 days after onset; ④ 18 to 71 years old; ⑤ Montreal Cognitive Assessment (MoCA) score <26; ⑥ no complications of neurodegenerative diseases; ⑦ no history of mental illness; ⑧ providing written informed consent. Exclusion criteria were: ① cognitive impairment before stroke; ② taking medications that may affect the assessment results; ③ severe language and / or cognitive impairment that prevented the subject from completing the assessment; ③ craniocerebral defect; ③ history of alcoholism.

[0103] 2. Evaluation process

[0104] 2.1 The task paradigm used in the present invention is a digital flash task. Specifically, a computer is used to present stimuli and collect participants' responses. Numbers 0-9 are randomly displayed on the computer screen, and number 7 is the target number. The stimulus is displayed in white on a black background in the center of the computer screen. The stimulus is presented at a constant rate of one every 1000 milliseconds, with a total of 24 target stimuli and 96 non-target stimuli. The subjects sat in comfortable seats. The subjects were asked to press the space bar with the index finger of the healthy side as soon as possible after seeing the number 7. The task was set to last for 5 minutes, and the changes in the Δ[oxy-Hb] signal were monitored during the 30-second baseline period before and after the task and during the task.

[0105] The digital flash task parameters used in the present invention have the following characteristics: First, the task cycle of the present invention is relatively long. A task cycle of 300s is used to examine the detailed time course of the frontal lobe oxygenated hemoglobin response. From the time course of changes in blood oxygen concentration, it can be seen that there is a continuous difference between the patients in the right hemisphere damage group and the control group throughout the task process. Second, the digital flash task used in the present invention is simple and easy to understand, and has a low difficulty. Compared with other cognitive activation tasks that require higher memory function, PSCI patients can also quickly adapt, master and cooperate to complete. Third, the digital flash task is more suitable for individual frontal lobe cognitive activation because it requires minimal motor response (space bar press) and does not require eye movement. This embodiment shows that the application of a simple cognitive activation task for a long time (300 seconds) can induce compensatory activation in PSCI patients with right hemisphere damage, and can also provide more sufficient and detailed time course characteristics of frontal lobe blood oxygen response.

[0106] 2.2 Use a near-infrared brain functional imaging system to collect near-infrared brain functional imaging data from the subjects, and obtain reliable near-infrared brain functional imaging data based on behavioral data. Specifically, the present invention uses the research-type near-infrared brain functional imaging system Light-NIRS (Shimadzu Enterprise Management (China) Co., Ltd. / Shimadzu (Hong Kong) Co., Ltd.), which has three wavelengths of near-infrared light, namely 780, 805 and 830nm. The sampling rate is 13.33Hz. 8 emitters and 8 receivers are arranged in a 2×8 array detection group. The distance between each emitter and receiver pair is 3cm. In the present invention, the midpoint of the corresponding emitter-receiver pair is defined as the test channel. According to the international 10-20 electrode configuration method, the bottom probe is placed along the Fp1-Fp2 line, and Fpz is located at the midpoint of the bottom probe connection line, ensuring that the probe group covers the left prefrontal cortex (left PFC) and the right prefrontal cortex (right PFC) Two areas of interest; the near-infrared brain functional imaging data corresponding to the coefficient of variation greater than 15% are discarded, and the rest are credible near-infrared brain functional imaging data. In addition, in this embodiment, 12 of the 22 channels can be included in the analysis of data, which is defined as valid data and can be used for subsequent analysis.

[0107] 2.3 Preprocessing of reliable near-infrared brain functional imaging data; Specifically, the present invention preprocesses near-infrared data based on the MATLAB homers2 toolkit, first converting the raw data light intensity into optical density, secondly, identifying data artifacts and marking bad channels through manual visual inspection, correcting data artifacts using automatic artifact recognition and homer2 interpolation correction methods, converting optical density into blood oxygen concentration, and using homer2 (0.003-0.1Hz) bandpass filter to remove physiological noise in the data, such as heartbeat (1-1.5Hz), breathing (0.2-0.3Hz), Mayer wave (about 0.1Hz), low-frequency physiological fluctuations (less than 0.01Hz), and high-frequency machine noise. During data processing, channels 4 and 19 located in the center are removed. The left PFC is covered by channels 5, 6, 7, 12, 13, 14, 15, 20, 21, and 22. The right PFC is covered by channels 1, 2, 3, 8, 9, 10, 11, 16, 17, and 18.

[0108] Since changes in oxygenated hemoglobin (oxy-Hb) have been shown to be more sensitive and reliable to local cerebral blood flow than deoxygenated hemoglobin (deoxy-Hb), in this embodiment, only changes in oxy-Hb are analyzed. The last 10 seconds of the baseline period before the task is taken as the baseline, and a baseline correction is performed. According to the position of the cerebral hemisphere to which the test channel belongs, the channels of the left PFC and the right PFC are superimposed and averaged, and the integral value and the center of gravity value are calculated on this basis. According to the behavioral data, integral value, and center of gravity value indicators of the subjects, the sustained attention function of the subjects is evaluated.

[0109] 3. Evaluation results

[0110] Behavioral results

[0111] The behavioral data with coefficient of variation less than 15% in the digit flash task were retained.During the digit flash task test, the accuracy and hit rate of the right hemisphere injury group were significantly lower than those of the control group (P<0.05), and the average reaction time was significantly higher than that of the control group (P<0.05).

[0112] Center of gravity and integral value

[0113] Compared with the control group, the right PFC center of gravity value in the right hemisphere injury group was higher than that in the control group, which was statistically significant. The mean of the bilateral PFC integral value in the right hemisphere injury group was higher than that in the control group, but there was no significant difference. As shown in Table 2;

[0114] Table 2 Comparison of integral values ​​and center of gravity values ​​between the right hemisphere injury group and the control group

[0115]

[0116] Table 2 shows the activation of the bilateral prefrontal cortex in PSCI patients with right hemisphere damage and the control group during the digital flash task. Compared with the control group, the bilateral prefrontal cortex activation increased in PSCI patients with right hemisphere damage, and the activation speed of the affected PFC in PSCI patients with right hemisphere damage was slower than that in the control group.

[0117] The PFC compensation speed of the right hemisphere injury group was slower and less efficient, indicating that the speed and quality of neural processing on the affected side of the RHD group decreased. Therefore, it is necessary to further recruit the healthy PFC to compensate for the reduction in processing efficiency. Therefore, when discussing the changes in brain activation in PSCI patients caused by cognitive training, it is necessary to discuss them separately according to the injured and non-injured brains. For the injured brain, cognitive training may lead to reduced activation of the injured brain due to improved processing efficiency, while increasing additional recruitment of the healthy hemisphere.

[0118] The differences in prefrontal responses between the control group and the right hemisphere damage group during the digital flash task test obtained in this example can be used in future clinical settings to assist in identifying PSCI patients with right cerebral hemisphere damage, further explain the causes of the behavioral manifestations from the perspective of cerebral cortical function, and provide a scientific basis for formulating personalized training programs.

[0119] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of the present invention to be protected is defined by the attached claims and their equivalents.

Claims

1. A sustained attention evaluation method based on near-infrared brain functional imaging, characterized in that: S1. Determine the brain function task paradigm and set the task specific parameters; S2. Determine the anatomical partition of the subject's cerebral cortex corresponding to the test channel, and use a near-infrared brain functional imaging device to monitor the brain function state of the subject during the task of step S1, and obtain near-infrared brain functional imaging data and behavioral data during the task; S3. judging the near-infrared brain function imaging data to obtain reliable near-infrared brain function imaging data; S4. Preprocess the reliable near-infrared brain functional imaging data; S5. Based on the processed near-infrared brain function imaging data, the near-infrared brain function imaging data of the same category of test channels are superimposed and averaged to calculate the corresponding integral value and center of gravity value; S6. Using the behavioral data in step S2 and the integral value and center of gravity value in step S5 as indicators, evaluate the subject's sustained attention function status.

2. The method for evaluating sustained attention based on near-infrared brain functional imaging according to claim 1, characterized in that: In step S1, the brain function task paradigm is a digit flash task.

3. The method for evaluating sustained attention based on near-infrared brain functional imaging according to claim 1, characterized in that: In step S2, the process of determining the anatomical partition of the subject's cerebral cortex corresponding to the test channel of the near-infrared brain functional imaging device includes: S201. Using the position coordinates of the skull reference point, establish a correspondence between the subject and the skull in a certain individual R1 structural head MRI image in the database, and based on the correspondence, map the positions of all the transmitters, receivers, and each test channel on the skull in the individual R1 head structural MRI image; S202. Based on the individual R1 head structural MRI image and the virtually placed optode positions, estimate the corresponding cortical positions and MNI coordinates of all test channels in the individual R1 head structural MRI image using a balloon inflation algorithm; S203. The remaining individuals R2-R in the database n After all steps S201 and S202 are repeated, each test channel obtains the same number of corresponding cortical positions and their MNI coordinates as the number of individuals in the database; S204. Obtain a final MNI coordinate based on the average of several MNI coordinates, and determine the anatomical partition of the cerebral cortex corresponding to the test channel according to the final MNI coordinate.

4. The method for evaluating sustained attention based on near-infrared brain functional imaging according to claim 1, characterized in that: In step S2, the behavioral data include: number of positive responses, number of incorrect responses, CANCEL, accuracy rate, hit rate, average reaction time, standard deviation and coefficient of variation.

5. The method for evaluating sustained attention based on near-infrared brain functional imaging according to claim 1, characterized in that: In step S3, the process of obtaining credible near-infrared brain function imaging data includes discarding the near-infrared brain function imaging data corresponding to the coefficient of variation in the behavioral data being greater than 15%.

6. The method for evaluating sustained attention based on near-infrared brain functional imaging according to claim 1, characterized in that: In step S4, the process of preprocessing the reliable near-infrared brain functional imaging data includes: S401. discarding the test channels in the reliable near-infrared brain functional imaging data with obvious motion artifacts and poor signal quality; S402. Based on the modified Beer-Lambert law, the original light intensity data is converted into relative change data of cerebral blood oxygen concentration, and the relative change data of cerebral blood oxygen concentration includes: oxygenated hemoglobin concentration Δ[oxy-Hb] signal, deoxygenated hemoglobin concentration Δ[deoxy-Hb] signal and total oxygenated hemoglobin concentration Δ[total-Hb] signal; S403. Use a 0.1 Hz low-pass filter to remove the noise components with obvious periodic fluctuation characteristics in the near-infrared brain functional imaging data; S404. Use the Savitzky-Golay method to smooth the data once, and the number of smoothing points is set to 5; S405. Take the average value of the relative change signal of the blood oxygen concentration within 15 seconds before the start of the task and perform baseline correction.

7. The method for evaluating sustained attention based on near-infrared brain functional imaging according to claim 1, characterized in that: In step S5, the process of calculating the integral value and the center of gravity value includes: S501. Based on the pre-processed near-infrared brain functional imaging data, according to the probabilistic registration result of the positioning data, the test channels with good quality are divided into different categories according to their Brodmann partitions and functions, and then the data of the test channels of the same category are superimposed and averaged; S502. Based on the superimposed and averaged data, calculate the corresponding integral value and center of gravity value.

8. The method for evaluating sustained attention based on near-infrared brain functional imaging according to claim 7, characterized in that: In step S502, the integral value calculation formula is: Where f(t) represents the signal function of the relative change of oxygenated hemoglobin concentration, t represents time, and t taskstart Represents the time when the task starts, t taskend represents the time point when the task ends, and d represents the differential.

9. The method for evaluating sustained attention based on near-infrared brain functional imaging according to claim 7, characterized in that: In step S502, the center of gravity value calculation process includes: S502-1. Calculate the total area of ​​the positive change region under the fNIRS positive signal during the entire test period; S502-2. Set a threshold value, which is 1 / 2 of the total area of ​​the positive change region; S502-3. From the start time of the test, gradually increase the time t, and calculate the area under the curve of the fNIRS positive signal up to the current time t. When the cumulative area is greater than or equal to the threshold, the time point is the centroid value, and its expression is: Where f(t) represents the signal function of the relative change of oxygenated hemoglobin concentration, and when f(t)>0, t represents time, t teststart Represents the time point when the test starts, t testend Represents the time point when the test ends, t centroidvalue is the time point corresponding to the centroid value.

Images