Method, equipment and program product for evaluating cognitive toughness

Abstract

The invention relates to the field of intelligent medical treatment, in particular to a method, equipment and a program product for evaluating cognitive toughness. Comprising the following steps: acquiring training data and test data for executing 5-CSRTT; calculating a comprehensive cognitive toughness index based on the training data and the test data; the comprehensive cognitive toughness index is a comprehensive index of different dimension capability values in the face of cognitive challenges, and the dimensions comprise any two or more of the following: basic performance, stability, volatility, restoring force and persistence; and performing cognitive toughness evaluation based on the comprehensive cognitive toughness index, when the comprehensive cognitive toughness index is greater than a first preset threshold value, determining that the toughness is high, when the comprehensive cognitive toughness index is equal to a second preset threshold value, determining that the toughness is medium, and when the comprehensive cognitive toughness index is smaller than a third preset threshold value, determining that the toughness is low. The application has good clinical value.

Description

Technical Field

[0001] This application relates to the field of smart healthcare, specifically to a method, device, program product, and computer-readable storage medium for assessing cognitive resilience. Background Technology

[0002] Significant individual differences exist in cognitive functions (such as attention, learning and memory, and executive function) in both humans and model animals (e.g., mice). These differences are not simply a matter of "good" or "bad," but rather form the basis of behavioral diversity. In healthy individuals, attention levels also exhibit a continuous distribution, with variations correlated with academic performance, work performance, and even quality of life. While individual differences in cognitive abilities have been widely observed in mice, an important behavioral model, these differences are often "averaged" in data analysis, and the underlying behavioral and neural mechanisms remain largely unexplored. The five consecutive reaction time tasks are a classic paradigm for measuring attention and impulsive behavior in rodents. This experimental paradigm requires animals to detect brief flashes of light appearing in a pseudo-random order at one of five spatial locations in a large number of tests. In this task, animals undergo approximately 30-40 days of daily training, during which they gradually learn to react to the correct hole within a specified time. In this task, experimental animals may exhibit four different types of responses, each corresponding to a different cognitive state: (i) they can respond correctly to a briefly presented stimulus after a long waiting time to obtain a reward (correct response), which requires a high level of attention and impulse control; (ii) they may also respond impulsively before the cue signal appears (premature response, indicating weakened impulse control); (iii) they may choose to respond to an unprovided hole (incorrect response, indicating decreased attention control); and (iv) they may not respond at all (missed response, indicating decreased task engagement or lack of attention). Previous studies have typically used the 5-CSRSTT experiment to test animal attention, merely focusing on superficial observations and lacking detailed analysis to uncover the more fundamental behavioral traits driving individual differences in "high / low attention." Furthermore, most studies, including pharmacological or genetic intervention studies, primarily focus on changes in the population average effect. Summary of the Invention

[0003] To address the above problems, this invention provides a method for assessing cognitive resilience, specifically including: Obtain training and test data for performing 5-CSRTT; the training data is the data of the sample performing 5-CSRTT in the first time period, and the test data is the data of the same sample performing 5-CSRTT in the second time period; The comprehensive cognitive resilience index is calculated based on the training and testing data. The comprehensive cognitive resilience index is a comprehensive index of different dimensions of ability values ​​when facing cognitive challenges. The dimensions include any two or more of the following: basic performance, stability, volatility, resilience, and endurance. Cognitive resilience is assessed based on the comprehensive cognitive resilience index. When the comprehensive cognitive resilience index is greater than a first preset threshold, it is determined to be high resilience; when the comprehensive cognitive resilience index is equal to a second preset threshold, it is determined to be medium resilience; and when the comprehensive cognitive resilience index is less than a third preset threshold, it is determined to be low resilience.

[0004] Optionally, the baseline performance is evaluated by Averaged accuracy and / or Averaged correct; Averaged accuracy refers to the average accuracy rate when completing various trials in the face of cognitive challenges, and Averaged correct refers to the average correctness rate when completing various trials in the face of cognitive challenges. Optionally, the averaged accuracy is obtained by calculating the average accuracy of the training data or the test data; Optionally, the average accuracy of the training data includes rank average accuracy and global average accuracy. The rank average accuracy is calculated by averaging the accuracy of the training data at any one rank, and the global average accuracy is calculated by averaging the accuracy of all the training data. Optionally, the Averaged correctness is obtained by calculating the average accuracy of the training data or the test data; Optionally, the average accuracy of the training data includes the rank average accuracy and the global average accuracy. The rank average accuracy is calculated by averaging the accuracy of the training data at any one rank, and the global average accuracy is calculated by averaging the accuracy of all the training data. Optionally, the training data includes six levels, with the difficulty of each level increasing sequentially.

[0005] Optionally, the stability is evaluated by the accuracy ratio and / or the correct ratio; the accuracy ratio refers to the ratio of the accuracy of any two levels of training data under cognitive challenges or the ratio of the accuracy of any level of training to the accuracy of testing; the correct ratio refers to the ratio of the accuracy of any two levels of training data under cognitive challenges or the ratio of the accuracy of any level of training to the accuracy of testing. Optionally, the accuracy ratio is calculated by comparing the sixth-level accuracy during the training phase with the accuracy during the testing phase; the correct ratio is calculated by comparing the sixth-level accuracy during the training phase with the accuracy during the testing phase.

[0006] Optionally, the volatility is evaluated using Accuracy CV and / or Correct CV; Accuracy CV refers to the CV value of accuracy between any two levels of training data under cognitive challenges, or the CV value of accuracy between any level of training and testing; Correct CV refers to the CV value of accuracy between any two levels of training data under cognitive challenges, or the CV value of accuracy between any level of training and testing.

[0007] Optionally, the resilience is evaluated by post-error accuracy, which refers to the proportion of correct trials appearing in L consecutive trials after erroneous trials appearing in training data and / or test data under cognitive challenges, where L is a natural number greater than 1. Optionally, the post-error accuracy is obtained by calculating the proportion of correct trials appearing in three consecutive trials after an erroneous trial appears in the training data and / or test data; Optionally, the post-error accuracy calculated using training data includes rank-based post-error accuracy and global post-error accuracy. The rank-based post-error accuracy is calculated using training data of any rank, and the global post-error accuracy is calculated using all training data. Optionally, the endurance is evaluated by the number of sequences of the longest consecutive correct trial, which is obtained by calculating the number of the longest consecutive correct trials in the training data and / or test data under the condition of facing a cognitive challenge; Optionally, the number of sequences of the longest consecutive correct trial, when calculated using training data, includes the number of sequences of the longest consecutive correct trial at the level and the number of sequences of the global longest consecutive correct trial. The number of sequences of the longest consecutive correct trial at the level is calculated using training data at any level, while the number of sequences of the global longest consecutive correct trial is calculated using all training data.

[0008] Optionally, the comprehensive cognitive resilience index calculates different dimension capability values ​​using training data and / or test data, and then compares the different dimension capability values ​​with preset thresholds for each dimension to obtain an evaluation result of high resilience, medium resilience, or low resilience. Optionally, the comprehensive cognitive resilience index calculates capability values ​​for basic performance, stability, volatility, resilience, and endurance dimensions using training data and / or test data. Then, it compares basic performance, stability, volatility, resilience, and endurance with corresponding preset thresholds to obtain first evaluation results for the five dimensions. Based on the first evaluation results of the five dimensions, a comprehensive judgment is made to obtain an evaluation result of high resilience, medium resilience, or low resilience. The comprehensive judgment is evaluated by the proportion of high resilience, medium resilience, or low resilience.

[0009] Optionally, the process of constructing the comprehensive cognitive resilience index is as follows: Obtain training and / or test data for performing 5-CSRTT; Calculate behavioral characteristics based on the training data and / or test data; Hierarchical clustering is performed on the behavioral features to calculate the correlation of the features; Based on the correlation, feature filtering is performed to obtain the filtered behavioral features; The selected behavioral features are normalized to obtain normalized behavioral features, the mean of the normalized behavioral features is calculated to obtain mean behavioral features, and the mean behavioral features are normalized to obtain the comprehensive cognitive resilience index. Optionally, the behavioral characteristics include various indicators in the dimensions of basic performance, stability, volatility, resilience, and endurance.

[0010] The purpose of this invention is to provide a computer program product that includes a computer program or instructions, which are executed by a processor to implement the above-described method for assessing cognitive resilience.

[0011] The purpose of this invention is to provide a computer device comprising a memory, a processor, and a computer program or instructions stored in the memory, wherein the computer program or instructions are executed by the processor to implement the above-described method for assessing cognitive resilience.

[0012] The purpose of this invention is to provide a computer-readable storage medium having a computer program or instructions stored thereon, which are executed by a processor to implement the above-described method for assessing cognitive resilience.

[0013] Advantages of this invention: 1. Current methods of attention analysis using the 5-CSRSTT approach only scratch the surface, lacking detailed analysis to uncover the more fundamental behavioral traits underlying individual differences in "high / low attention" and to address changes in the average effect of the primary focus group. Furthermore, this invention deconstructs different behavioral characteristics and incorporates sequential distribution behavioral analysis to discover the behavioral trait of "cognitive resilience" hidden behind the stratification of attention abilities. It further employs hierarchical clustering and Z-score methods to construct a comprehensive "cognitive resilience" assessment system. This system evaluates cognitive resilience through a comprehensive cognitive resilience assessment index, which comprises five dimensions and multiple indicators, providing a more comprehensive and stable capture of this trait.

[0014] 2. Addressing the differences in attention among individuals, this invention proposes behavioral characteristics or indicators across various dimensions, including volatility indicators, stability indicators, resilience indicators, persistence indicators, and basic performance indicators. The novel behavioral indicators developed in this invention provide new tools for the field. These indicators may be more sensitive and interpretive than traditional indicators, capable of detecting subtle effects that are not apparent at the group average level.

[0015] 3. The technical solution proposed in this invention will also provide a foundation for precision medicine: understanding the biological basis of cognitive traits (such as resilience) within healthy populations is the first step towards "precision mental health." In the future, this will help us understand why different people respond differently to the same environmental stress or cognitive training. Furthermore, the behavioral trait benchmarks established in healthy animals in this study lay a solid foundation for future research into the deficit mechanisms of related traits in models of mental disorders. It can also guide researchers to better design and interpret behavioral experiments; for example, in drug screening, innate cognitive traits in animals can be considered to obtain more accurate and reproducible results. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of a method for assessing cognitive resilience provided in an embodiment of the present invention; Figure 2 A schematic diagram of a system for assessing cognitive resilience provided in an embodiment of the present invention; Figure 3 A schematic diagram of a computer device provided in an embodiment of the present invention; Figure 4 The following diagram illustrates the individual differences exhibited by mice in four independent batches during the training phases provided in this embodiment of the invention: A represents the cumulative total time spent by all batches of mice at different stages of the training phase, with different colors representing different stages, and batches represented by numbers at the beginning of the mouse number. B represents a violin plot of the total time spent by mice during the training phase. CE represents the learning curve distribution diagrams of the mice's Accuracy, Omission, and Premature. Figure 5 This diagram illustrates the individual differences exhibited by mice in four independent batches during the testing phases provided in this embodiment of the invention; A represents the percentage of different types of trials in all batches of mice during the testing phase. B represents the statistical distribution of relevant behavioral indicators in the mouse testing phases. Figure 6 The following diagram illustrates the grouping indicators and basic behavioral differences of mouse attention ability provided in this embodiment of the invention; A represents the percentage of different types of trials in all batches of mice during the testing phase; B represents the statistical distribution of relevant behavioral indicators in the mouse testing phase. Figure 7 The comparison of behavioral characteristics hidden behind the stratification of mouse attention ability provided in the embodiments of the present invention; AB represents the comparison of learning ability; C represents the comparison of stability and volatility indicators; D represents the comparison of sequence analysis; and E represents the relevant indicators derived from sequence analysis. Figure 8 The construction and behavioral characteristic verification of the mouse cognitive resilience index provided in the embodiments of the present invention. Detailed Implementation

[0018] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0019] In some of the processes described in the specification, claims, and accompanying drawings of this invention, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as S101, S102, etc., are merely used to distinguish different operations and do not represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0020] Figure 1The schematic diagram of the method for assessing cognitive resilience provided in this embodiment of the invention specifically includes: S1: Obtain the training and test data for performing 5-CSRTT; the training data is the data of the sample performing 5-CSRTT in the first time period, and the test data is the data of the same sample performing 5-CSRTT in the second time period; In one embodiment, the training data for performing 5-CSRTT can come from laboratory animals such as mice, rats, and rabbits; Alternatively, the training data for 5-CSRTT can also come from primates or humans.

[0021] In one embodiment, the first time period precedes the second time period, and the duration of the first time period is not equal to the duration of the second time period.

[0022] In one embodiment, this invention deconstructs "attention" from a single-dimensional performance into a more complex construct supported by multiple behavioral traits, such as resilience. This tells us that two mice with the same accuracy may have completely different behavioral pathways and intrinsic traits to achieve this performance. Furthermore, we link "cognitive resilience" to basic attention performance, providing a novel behavioral framework for understanding why some individuals can maintain stable performance under distraction, fatigue, or time pressure.

[0023] In one embodiment, the classic 5-CSRTT behavioral metric is: 1. Accuracy is the most important indicator for reflecting attention, and it is calculated as: [correct responses / (correct responses + incorrect responses)] × 100. 2. Omissions can reflect attention, but they can also be affected by factors such as motivation and motor skills. The calculation formula is: =omissions / (omissions + correct responses + incorrect responses)] ×100.

[0024] 3. Premature is the main indicator of impulse control ability. It can be calculated as Premature / totaltrials or by counting the number of Prematures. This invention uses the Premature percentage.

[0025] 4. "Correct" represents the accuracy rate, reflecting the overall performance of mice in the 5-CSRTT. Calculation method: Correct / Total Trials × 100.

[0026] In one specific embodiment, this invention compiled and organized 5-CSRTT behavioral training data from four batches of healthy male mice. These data came from four different training batches. Batch 1 and Batch 2 used the same equipment and the same brand and size of food particles. Batch 3 and Batch 4 used the same size food particles, but different brands of training equipment. We first analyzed and compared the data during the mouse training process. We found that regardless of the batch, the total training time for mice through stages 1-6 showed significant individual differences (this indicator reflects the mouse's learning ability and efficiency). Figure 4 As shown. After all mice passed the stage 6 pass criteria, they underwent training for the testing phase. We found that even when all mice used the same pass criteria, they still consistently exhibited significant individual differences when entering the testing phase. This included the proportion of mice in different types of trials and comparisons of different behavioral traits during the testing phase, such as... Figure 5 As shown.

[0027] S2: Calculate the comprehensive cognitive resilience index based on the training data and test data; the comprehensive cognitive resilience index is a comprehensive index of different dimensions of ability values ​​under cognitive challenges, and the dimensions include any two or more of the following: basic performance, stability, volatility, resilience, and endurance; In one embodiment, the baseline performance is evaluated by Averaged accuracy and / or Averaged correct; Averaged accuracy refers to the average accuracy rate when completing various trials in the face of cognitive challenges, and Averaged correct refers to the average correctness rate when completing various trials in the face of cognitive challenges.

[0028] In one embodiment, the averaged accuracy is obtained by calculating the average accuracy of the training data or the test data.

[0029] In one embodiment, calculating the average accuracy of the training data includes rank average accuracy and global average accuracy. The rank average accuracy is calculated by averaging the accuracy of training data at any one rank, and the global average accuracy is calculated by averaging the accuracy of all training data.

[0030] In one embodiment, the Averaged correctness is obtained by calculating the average accuracy of the training data or the test data.

[0031] In one embodiment, calculating the average accuracy of the training data includes rank-based average accuracy and global average accuracy. The rank-based average accuracy is calculated by averaging the accuracy of the training data at any one rank, while the global average accuracy is calculated by averaging the accuracy of all training data. Evaluation is performed using the rank-based average accuracy and / or the global average accuracy.

[0032] In one embodiment, the training data includes six levels, with the difficulty of each level increasing sequentially.

[0033] In one embodiment, the stability is evaluated by the accuracy ratio and / or the correct ratio; the accuracy ratio refers to the ratio of the accuracy of any two levels of training data under cognitive challenges or the ratio of the accuracy of any level of the training phase to the accuracy of the testing phase; the correct ratio refers to the ratio of the accuracy of any two levels of training data under cognitive challenges or the ratio of the accuracy of any level of the training phase to the accuracy of the testing phase.

[0034] In one embodiment, the accuracy ratio is calculated by comparing the sixth-level accuracy during the training phase with the accuracy during the testing phase; the correct ratio is calculated by comparing the sixth-level accuracy during the training phase with the accuracy during the testing phase.

[0035] In one embodiment, the volatility is evaluated using Accuracy CV and / or Correct CV; the Accuracy CV refers to the CV value of accuracy at any two levels of training data under cognitive challenges, or the CV value of accuracy at any level of training and testing; the Correct CV refers to the CV value of accuracy at any two levels of training data under cognitive challenges, or the CV value of accuracy at any level of training and testing.

[0036] In one embodiment, the resilience is evaluated by post-error accuracy, which refers to the proportion of correct trials occurring in L consecutive trials after erroneous trials occur in training and / or testing data under cognitive challenges, where L is a natural number greater than 1.

[0037] In one embodiment, the post-error accuracy is obtained by calculating the proportion of correct trials occurring in three consecutive trials after an erroneous trial occurs in the training data and / or test data.

[0038] In one embodiment, the post-error accuracy, calculated using training data, includes rank-based post-error accuracy and global post-error accuracy. The rank-based post-error accuracy is calculated using training data of any rank, while the global post-error accuracy is calculated using all training data. Evaluation is performed using the rank-based post-error accuracy and / or the global post-error accuracy.

[0039] In one embodiment, the endurance is evaluated by the number of sequences of the longest consecutive correct trial, which is obtained by calculating the number of the longest consecutive correct trials in training and / or test data in the face of cognitive challenges.

[0040] In one embodiment, the number of sequences in the longest consecutive correct trial, calculated using training data, includes the number of sequences in the longest consecutive correct trial at the rank level and the number of sequences in the longest consecutive correct trial globally. The number of sequences in the longest consecutive correct trial at the rank level is calculated using training data at any rank, while the number of sequences in the longest consecutive correct trial globally is calculated using all training data. Evaluation is performed using the number of sequences in the longest consecutive correct trial at the rank level and / or the number of sequences in the longest consecutive correct trial globally.

[0041] In one embodiment, the comprehensive cognitive resilience index calculates capability values ​​for different dimensions using training data and / or test data, and then compares these capability values ​​with preset thresholds for each dimension to obtain an evaluation result of high resilience, medium resilience, or low resilience.

[0042] In one embodiment, the comprehensive cognitive resilience index calculates capability values ​​for basic performance, stability, volatility, resilience, and endurance dimensions using training data and / or test data. Then, it compares each of these dimensions with a corresponding preset threshold to obtain a first evaluation result for each of the five dimensions. Based on these first evaluation results, a comprehensive judgment is made to determine whether the resilience is high, medium, or low. This comprehensive judgment is based on the proportion of high, medium, or low resilience.

[0043] In one embodiment, the process of constructing the comprehensive cognitive resilience index is as follows: Obtain training and / or test data for performing 5-CSRTT; Calculate behavioral characteristics based on the training data and / or test data; Hierarchical clustering is performed on the behavioral features to calculate the correlation of the features; Based on the correlation, feature filtering is performed to obtain the filtered behavioral features; The selected behavioral features are normalized to obtain normalized behavioral features, the mean of the normalized behavioral features is calculated to obtain mean behavioral features, and the mean behavioral features are normalized to obtain the comprehensive cognitive resilience index. Optionally, the behavioral characteristics include various indicators in the dimensions of basic performance, stability, volatility, resilience, and endurance.

[0044] The behavioral characteristics include indicators from five dimensions: basic performance, stability, volatility, resilience, and endurance. After screening the behavioral characteristics, the selected characteristics are obtained. The selected characteristics include five dimensions. The screening process involves selecting indicators from each of the five dimensions. The selected indicators from the five dimensions are then normalized, mean calculated, and normalized to obtain the values ​​of each indicator in the five dimensions. These values ​​are used as a comprehensive cognitive resilience index, which is further used for resilience assessment to obtain the evaluation results.

[0045] In one specific embodiment, the behavioral characteristics underlying the stratification of mouse attention abilities are analyzed: Since the 5-CSRTT is a classic paradigm for detecting attention, and accuracy is the gold standard for evaluating attention, and performance in the stage 6 training phase is generally considered the baseline attention ability of mice, and we change the training conditions in the testing phase, we used the average accuracy of the last three days of stage 6 and the three days of the testing phase as the standard for evaluating the mice's attention ability to assess differences in attention ability. We sorted the above averages in descending order, selecting animals with a value greater than Q1 as animals with attention advantage, animals with a value equal to the median as animals with average attention, and animals with a value less than Q3 as animals with poor attention. To understand whether the behavioral performance of these high, medium, and low attention mice was affected by factors such as motivation and decision-making ability, we further compared the mice's motivation level (mainly using behavioral indicators such as omission, reward response latency, and magazine entries) and decision-making ability or reaction speed (mainly using two behavioral indicators: latency of correct response and latency of incorrect response). Through statistical analysis, we found that individuals with differences in attention ability did not show significant differences in motivation and decision-making ability in the 5-CSRTT. Figure 6 As shown in the figure, the mouse groups we obtained are typical groups with differences in attention ability.

[0046] We further compared the learning abilities of mice during the training process and found no significant differences in learning ability (the learning ability indicators here surpass the existing indicators of the traditional 5-CSRTT). We further calculated the volatility of the mice's performance during the testing phase and innovatively introduced two indicators: CV (Continuous Characteristic) and accuracy ratio and correct ratio. We found that the high-attention group had lower CV and higher accuracy ratio and correct ratio. We further systematically analyzed the sequence distribution characteristics of different types of trials during the testing phase, extracting the number of longest correct segments, the accuracy rate after an error (the proportion of correct trials in three consecutive trials after an error), and the slope of omission at different stages (the 100 trials in the testing phase were divided into four equal parts, the omission in each part was calculated, and the omission slope at different stages was obtained). We found that the longest correct segment length and the accuracy rate after an error were significantly better in animals with high attention than in animals with low attention. These behavioral aspects with significant individual differences are precisely the manifestation of "cognitive resilience". Figure 7 (As shown).

[0047] In one specific embodiment, 5-CSRTT is commonly used to detect attention and impulsive behavior in animal disease models. A few studies utilize data from its training phase to assess learning ability. Indicators of learning ability are typically compared by plotting learning curves, including the slope of the learning curve; this is the most common method used in various fields to study learning ability. In this invention… Figure 7 The system compared the learning ability metrics in 5-CSRTT, including the learning curve ( Figure 7 The A metric includes three lines: Accuracy%, Omission%, and Premature%. The horizontal axis represents different training stages of varying difficulty, and the vertical axis represents the average Accuracy%, Omission%, and Premature% of the training stages. The Learning Slope is the slope of the learning curve; a steeper slope indicates stronger learning ability. In addition, I compared the training duration (T1-T3) for completing training stages 1-3. This metric is defined based on the learning curve and learning patterns. It reflects the ability during the rule-learning stage because the 5-CSRTT behavioral test requires experimental animals to undergo long-term training stages consisting of six rules of varying difficulty. In plotting the learning curve, we found that the curve became flat after training stage 4. Figure 7The A Accuracy% curve represents the time taken for mice to fully master the training rule by training phase 4. Therefore, the time taken for the animals to complete training phases 1-3 can be used to reflect the mice's rule learning speed. This invention also compared the total number of days (T1-T6) spent completing the entire training phase (stage 1-stage 6), which reflects the mice's learning speed and efficiency. Furthermore, this invention compared attention optimization speed (Δaccuracy / day) and impulse control maturity (Δpremature / day). The calculation for attention optimization speed (Δaccuracy / day) is: (Average Accuracy in stage 6 - Average Accuracy in stage 3) / Total number of days spent in stages 4-6, multiplied by 100. The calculation for impulse control maturity is: (Average maturity in stage 3 - Average maturity in stage 6) / Total number of days spent in stages 3-6, multiplied by 100. These two indicators respectively reflect the learning efficiency of different behavioral dimensions. Therefore... Figure 7 Parts A and B are an extension and comparison of learning abilities.

[0048] In one specific embodiment, CV is an abbreviation for Coefficient of Variation, which is a statistic that measures the degree of dispersion of a set of data and is typically used to compare the relative dispersion of different datasets.

[0049] The calculation method is: CV = (standard deviation / mean) × 100%; In this invention, the CV of Accuracy% and Correct% is calculated for the mouse training phase stage 6 (3 days) and the Test phase (3 days). This indicator reflects the variability of Accuracy% and Correct%.

[0050] In one specific embodiment, the accuracy ratio is the accuracy percentage, and the formula for calculating the accuracy ratio (T6 / T5) is = Accuracy% of training stage 6 (stage 6) / Accuracy of training stage 5 (stage 5). The accuracy ratio (Test / T6) = Accuracy% of the challenge test stage (Test) / Accuracy of training stage 6 (stage 6).

[0051] The correct ratio is the same as above, except that accuracy is replaced with correctness. These metrics reflect the stability of the animal's accuracy% and correctness% when faced with challenging tasks or more difficult tests.

[0052] In one specific embodiment, Figure 7 In D, this invention compares the proportion of different types of trials in experimental animals during the challenge test phase. Different types of trials reflect different behavioral states of the animals. Correct trials indicate that the experimental animal completed the experiment with high concentration and accuracy; incorrect trials indicate that the experimental animal's attention is scattered and the task was not completed accurately; missed trials indicate that the experimental animal's attention is not focused or its motivation is reduced, resulting in slower movement speed; impulsive trials indicate that the animal has not controlled its impulses well, leading to impulsive behavior. Analyzing the composition and changes of these behavioral components over time can better help us extract components of cognitive ability. This invention compares the number of the longest consecutive correct trials. This indicator is calculated as follows: if a correct trial occurs twice consecutively, it is counted as 2; if a correct trial occurs three times consecutively, it is counted as 3, and so on, to find the longest consecutive correct trial among all trials tested that day. This indicator reflects the persistence of attention (e.g., ...). Figure 7 As shown in E (left figure).

[0053] This invention also statistically analyzes the accuracy rate after errors ( Figure 7 The E (intermediate figure) is calculated as the percentage of correct trials appearing in three consecutive trials after an incorrect trial. This indicator reflects more the mouse's ability to quickly adjust its behavior after making an incorrect decision.

[0054] The slope of the change in the omission rate was also statistically analyzed. Figure 7 (See Figure E on the right). The calculation method is as follows: First, divide the 100 test trials into 4 equal parts, with 25 trials in each part. First, calculate the omisson% in each part, and then calculate the slope of the change in omisson% across the 4 parts. The change in omission rate can also comprehensively reflect the persistence of good cognitive abilities in experimental animals.

[0055] In one specific embodiment, based on the above findings, we further analyzed the correlation between these statistically significant behavioral characteristics using hierarchical clustering. Cluster analysis revealed that these behavioral characteristics mainly manifested in the following dimensions: stability, volatility, basic ability, post-error accuracy, and persistence. Based on this, we constructed a comprehensive cognitive resilience index using these five dimensions and eight behavioral indicators. Using the values ​​of the comprehensive cognitive resilience index as the grouping standard, we sorted the values ​​in descending order, selecting those greater than Q1 as the high-resilience group, those equal to the median as the medium-resilience group, and those less than Q3 as the low-cognitive-resilience group. The construction and behavioral characteristic verification of the mouse cognitive resilience index are as follows. Figure 8 As shown, Figure 8 AB diagrams are cluster analysis and network diagrams of correlations between behavioral parameters with statistical differences. Figure 8 C represents the constituent dimensions of the cognitive resilience index. Figure 8 D represents the feature distribution after grouping by cognitive resilience index (left), and the feature distribution of static grouping of attention ability (right). Figure 8 E represents the heatmap showing the correlation between the remaining behavioral characteristics and the comprehensive cognitive resilience index. After grouping by cognitive resilience, the distribution of the eight behavioral characteristics in different groups is consistent with that in the static attention ability group. We further analyzed the correlation between the comprehensive cognitive resilience index and all behavioral characteristics during the mouse training and testing processes, finding that the behavioral characteristics of the mice grouped by comprehensive cognitive resilience were consistent with those in the static attention ability group. This fully demonstrates that the cognitive resilience code is hidden behind the attention ability stratification.

[0056] S3: Based on the comprehensive cognitive resilience index, cognitive resilience is assessed. When the comprehensive cognitive resilience index is greater than the first preset threshold, it is determined to be high resilience; when the comprehensive cognitive resilience index is equal to the second preset threshold, it is determined to be medium resilience; and when the comprehensive cognitive resilience index is less than the third preset threshold, it is determined to be low resilience.

[0057] In one embodiment, the first preset threshold is the first quartile of Q1 in the cognitive resilience grouping experiment, the second preset threshold is the median, and the third preset threshold is the third quartile of Q3. Figure 7 As shown in D.

[0058] In one specific embodiment, the academic community lacks a unified definition and concept of cognitive resilience. In recent studies, cognitive resilience has been mentioned and defined in AD research as: the ability of an individual to maintain cognitive function better than expected despite neuropathological changes. The concept of cognitive resilience typically stems from the understanding and definition of resilience, which originally refers to the rebound of an object when subjected to external force. It is extended to describe the phenomenon of an individual's adaptation and development remaining good despite severe threats. To date, the academic community has not reached a unified understanding of the concept of cognitive resilience. However, based on the findings of our research, this invention summarizes cognitive resilience as: Cognitive resilience refers to an individual's (e.g., a mouse's) comprehensive ability to maintain, recover, and optimize its goal-oriented behavioral performance when faced with cognitive challenges. It is not a single indicator of "attention" or "impulsivity," but a multidimensional and stable behavioral trait, the core of which lies in the ability to maintain the efficient and stable operation of the cognitive system under inherent challenges such as time pressure, erroneous feedback, and task monotony.

[0059] In one specific embodiment, Figure 8This paper presents the construction and components of cognitive resilience proposed in this study: First, hierarchical clustering was used to analyze the correlations among statistically significant behavioral characteristics, including Averaged accuracy, Averaged correct, Accuracy CV, Accuracy ratio (Test / T6), Accuracy ratio (T6 / T5), Correct CV, Correct ratio (Test / T6), Longest sequence of correct trial, and post-incorrect accuracy. These indicators are all behavioral dimensions with statistically significant differences after grouping by attention ability. We further calculated the Pearson correlation coefficients between different behavioral indicators and, based on this, used hierarchical clustering analysis and graph theory to construct a network diagram of the relationships between the above behavioral indicators. We found that Averaged accuracy, Averaged correct, Accuracy ratio (Test / T6), and Correct ratio (Test / T6) belong to one class; Accuracy CV and Correct CV belong to another class; and Longest sequence of correct trial, post-incorrect accuracy, and Accuracy ratio (T6 / T5) belong to yet another class. Since Accuracy and Correctness represent different dimensions, and their Pearson correlation coefficients (r) do not show a high degree of linear correlation, we include both in the evaluation system. (Accuracy is the primary indicator of attention, calculated as [correct responses / (correct responses + incorrect responses)] × 100; Correctness is a comprehensive indicator reflecting the overall cognitive level of mice in the 5-CSRTT, calculated as [correct responses / (correct responses + incorrect responses + omission + premature)] × 100). Furthermore, the Accuracy ratio (T6 / T5) shows a weak correlation with other indicators, so we do not include this indicator.Based on my above introduction to the various indicators, we will include: Averaged accuracy, Averaged correct, Accuracy ratio (Test / T6), Correct ratio (Test / T6), Accuracy CV, Correct CV, Longest sequence of correct trial, and Post-incorrect accuracy. These indicators reflect five dimensions closely related to resilience: basic performance under cognitive challenges (Averaged accuracy and Averaged correct), stability under cognitive challenges (Accuracy ratio (Test / T6) and Correct ratio (Test / T6), volatility under cognitive challenges (Accuracy CV and Correct CV), resilience under cognitive challenges (Post-incorrect accuracy), and persistence of attention under cognitive challenges (Longest sequence of correct trial). Based on this, I constructed a comprehensive cognitive resilience index using these five dimensions and eight behavioral indicators.

[0060] Methodology for constructing the comprehensive cognitive resilience index: Since the data units for these behavioral dimensions are inconsistent and their values ​​differ, we first performed Z-Score normalization on the data for these behavioral dimensions. (Z-Score normalization is a commonly used data preprocessing method used to convert data into a standard normal distribution with a mean of 0 and a standard deviation of 1. This method can eliminate the dimensional differences between different features, allowing data to be compared on the same scale, thereby improving model performance and the fairness of the analysis. The formula for calculating the Z-Score is: Z = (X - μ) / σ). We first calculated the Z-Score values ​​for these 8 indicators separately, and then calculated the Averaged Z-Score values ​​for these 8 indicators, using these values ​​as the comprehensive cognitive resilience index. We used the comprehensive cognitive resilience index values ​​as the grouping standard, sorting the values ​​in descending order (the larger the value, the higher the "cognitive resilience" ability). Values ​​greater than the first quartile of Q1 were selected as the high-resilience group, values ​​equal to the median as the medium-resilience group, and values ​​less than the third quartile of Q3 as the low-cognitive resilience group.

[0061] The present invention also discloses a computer program product or system, including a computer program that, when executed by a processor, implements the above-described method steps for assessing cognitive resilience.

[0062] Figure 2 The system diagram for assessing cognitive resilience provided in this embodiment of the invention specifically includes: Acquisition Unit: Acquires training and test data for performing 5-CSRTT; Calculation unit: Calculates the comprehensive cognitive resilience index based on the training data and test data; the comprehensive cognitive resilience index is a comprehensive index of different dimensions of ability values ​​in the face of cognitive challenges, and the dimensions include any two or more of the following: basic performance, stability, volatility, resilience, and endurance; Evaluation Unit: Based on the comprehensive cognitive resilience index, cognitive resilience is evaluated. When the comprehensive cognitive resilience index is greater than the first preset threshold, it is determined to be high resilience; when the comprehensive cognitive resilience index is equal to the second preset threshold, it is determined to be medium resilience; and when the comprehensive cognitive resilience index is less than the third preset threshold, it is determined to be low resilience.

[0063] Figure 3 An embodiment of the present invention provides a schematic diagram of a computer device, specifically including: A memory and a processor; the memory is used to store program instructions; the processor is used to invoke the program instructions when any of the above-described methods for assessing cognitive resilience are executed.

[0064] The present invention also discloses a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, is any of the above-described methods for assessing cognitive resilience.

[0065] The verification results of this verification embodiment show that assigning inherent weights to indications can improve the performance of this method compared to the default settings. Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for example, the division of units is merely a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be electrical, mechanical, or other forms. The units described as separate components may or may not be physically separated; the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of this embodiment. Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units. Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. This program can be stored in a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0066] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0067] The computer device provided by the present invention has been described in detail above. For those skilled in the art, there will be changes in the specific implementation and application scope based on the ideas of the embodiments of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for assessing cognitive resilience, characterized in that, include: Obtain training and test data for performing 5-CSRTT; The training data consists of the data from the first time period when the sample is subjected to 5-CSRTT, and the test data consists of the data from the second time period when the same sample is subjected to 5-CSRTT. The comprehensive cognitive resilience index is calculated based on the training and testing data. The comprehensive cognitive resilience index is a comprehensive index of different dimensions of ability values ​​when facing cognitive challenges. The dimensions include any two or more of the following: basic performance, stability, volatility, resilience, and endurance. Cognitive resilience is assessed based on the comprehensive cognitive resilience index. When the comprehensive cognitive resilience index is greater than a first preset threshold, it is determined to be high resilience; when the comprehensive cognitive resilience index is equal to a second preset threshold, it is determined to be medium resilience; and when the comprehensive cognitive resilience index is less than a third preset threshold, it is determined to be low resilience.

2. The method for assessing cognitive resilience according to claim 1, characterized in that, The baseline performance is assessed by Averaged accuracy and / or Averaged correct; Averaged accuracy refers to the average accuracy rate when completing various trials in the face of cognitive challenges, and Averaged correct refers to the average correctness rate when completing various trials in the face of cognitive challenges. Optionally, the averaged accuracy is obtained by calculating the average accuracy of the training data or the test data; Optionally, the average accuracy of the training data includes rank average accuracy and global average accuracy. The rank average accuracy is calculated by averaging the accuracy of the training data at any one rank, and the global average accuracy is calculated by averaging the accuracy of all the training data. Optionally, the Averaged correctness is obtained by calculating the average accuracy of the training data or the test data; Optionally, the average accuracy of the training data includes the rank average accuracy and the global average accuracy. The rank average accuracy is calculated by averaging the accuracy of the training data at any one rank, and the global average accuracy is calculated by averaging the accuracy of all the training data. Optionally, the training data includes six levels, with the difficulty of each level increasing sequentially.

3. The method for assessing cognitive resilience according to claim 1, characterized in that, The stability is evaluated by the accuracy ratio and / or the correct ratio; the accuracy ratio is the ratio of the accuracy of any two levels of training data under cognitive challenge or the ratio of the accuracy of any level of training to the accuracy of testing; the correct ratio is the ratio of the accuracy of any two levels of training data under cognitive challenge or the ratio of the accuracy of any level of training to the accuracy of testing. Optionally, the accuracy ratio is calculated by comparing the sixth-level accuracy during the training phase with the accuracy during the testing phase; the correct ratio is calculated by comparing the sixth-level accuracy during the training phase with the accuracy during the testing phase.

4. The method for assessing cognitive resilience according to claim 1, characterized in that, The volatility is evaluated using Accuracy CV and / or Correct CV; Accuracy CV refers to the CV value of accuracy at any two levels of training data under cognitive challenges, or the CV value of accuracy at any level of training and testing; Correct CV refers to the CV value of accuracy at any two levels of training data under cognitive challenges, or the CV value of accuracy at any level of training and testing.

5. The method for assessing cognitive resilience according to claim 1, characterized in that, The resilience is evaluated by post-error accuracy, which refers to the proportion of correct trials appearing in L consecutive trials after erroneous trials appearing in training data and / or test data under cognitive challenges, where L is a natural number greater than 1. Optionally, the post-error accuracy is obtained by calculating the proportion of correct trials appearing in three consecutive trials after an erroneous trial appears in the training data and / or test data; Optionally, the post-error accuracy calculated using training data includes rank-based post-error accuracy and global post-error accuracy. The rank-based post-error accuracy is calculated using training data of any rank, and the global post-error accuracy is calculated using all training data. Optionally, the endurance is evaluated by the number of sequences of the longest consecutive correct trial, which is obtained by calculating the number of the longest consecutive correct trials in the training data and / or test data under the condition of facing a cognitive challenge; Optionally, the number of sequences of the longest consecutive correct trial, when calculated using training data, includes the number of sequences of the longest consecutive correct trial at the level and the number of sequences of the global longest consecutive correct trial. The number of sequences of the longest consecutive correct trial at the level is calculated using training data at any level, while the number of sequences of the global longest consecutive correct trial is calculated using all training data.

6. The method for assessing cognitive resilience according to claim 1, characterized in that, The comprehensive cognitive resilience index calculates capability values ​​for different dimensions using training data and / or test data, and then compares these capability values ​​with preset thresholds for each dimension to obtain an evaluation result of high resilience, medium resilience, or low resilience. Optionally, the comprehensive cognitive resilience index calculates capability values ​​for basic performance, stability, volatility, resilience, and endurance dimensions using training data and / or test data. Then, it compares basic performance, stability, volatility, resilience, and endurance with corresponding preset thresholds to obtain first evaluation results for the five dimensions. Based on the first evaluation results of the five dimensions, a comprehensive judgment is made to obtain an evaluation result of high resilience, medium resilience, or low resilience. The comprehensive judgment is evaluated by the proportion of high resilience, medium resilience, or low resilience.

7. The method for assessing cognitive resilience according to claim 1, characterized in that, The process of constructing the comprehensive cognitive resilience index is as follows: Obtain training and / or test data for performing 5-CSRTT; Calculate behavioral characteristics based on the training data and / or test data; Hierarchical clustering is performed on the behavioral features to calculate the correlation of the features; Based on the correlation, feature filtering is performed to obtain the filtered behavioral features; The selected behavioral features are normalized to obtain normalized behavioral features, the mean of the normalized behavioral features is calculated to obtain mean behavioral features, and the mean behavioral features are normalized to obtain the comprehensive cognitive resilience index. Optionally, the behavioral characteristics include various indicators in the dimensions of basic performance, stability, volatility, resilience, and endurance.

8. A computer program product comprising a computer program or instructions, characterized in that, The computer program or instructions are executed by a processor to implement the method for assessing cognitive resilience as described in any one of claims 1-7.

9. A computer device comprising a memory, a processor, and a computer program or instructions stored in the memory, characterized in that, The computer program or instructions are executed by a processor to implement the method for assessing cognitive resilience as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, The computer program or instructions are executed by a processor to implement the method for assessing cognitive resilience as described in any one of claims 1-7.

Images