Auxiliary assessment method and system for cognitive impairment based on big data

Abstract

The invention provides an auxiliary assessment method and system for cognitive impairment based on big data, relates to the technical field of diagnosis, and can significantly improve the accuracy and sensitivity of early screening of cognitive impairment and provide interpretable assessment results and personalized intervention suggestions on the premise of realizing automation and objectification of the assessment process, thereby improving the assessment efficiency. The method comprises the steps of obtaining dynamic data of a to-be-evaluated object, wherein the dynamic data is used for representing basic information and cognitive information of the to-be-evaluated object; determining a dynamic deviation degree with a health baseline model based on the dynamic data; performing cognitive impairment assessment operation based on the dynamic deviation degree to obtain a cognitive impairment risk level of the to-be-assessed object; based on the dynamic data and the dynamic deviation degree, a damaged area is determined, and the damaged area is an area where the brain cognitive impairment function of the to-be-evaluated object is damaged; and determining an intervention suggestion based on the cognitive disorder risk level and the damaged area of the to-be-evaluated object.

Description

Technical Field

[0001] This application relates to the field of diagnostic technology, and in particular to an auxiliary assessment method and system for cognitive impairment based on big data. Background Technology

[0002] With the accelerating aging of the global population, cognitive impairment caused by neurodegenerative diseases, such as Alzheimer's disease, is becoming increasingly serious, placing a heavy burden on families and society. Early detection and intervention of cognitive impairment are crucial for slowing disease progression and maintaining patients' quality of life.

[0003] In current technologies, the assessment of cognitive impairment mainly relies on traditional neuropsychological scales (such as the Mini-Mental State Examination (MMSE) and the Montreal Cognitive Assessment (MoCA)) and the subjective judgment of clinicians. These methods have the following limitations: First, the assessment results are easily influenced by factors such as the assessor's experience, the patient's current state, and their education level, making them highly subjective; second, the assessments are usually intermittent "snapshot" tests, making it difficult to capture dynamic changes and subtle decline trends in cognitive function, resulting in insufficient sensitivity for early screening; third, the assessment process requires the participation of professional medical personnel, is time-consuming, labor-intensive, and costly, and is difficult to apply to routine screening of large populations. In recent years, although some mobile application-based cognitive training and assessment tools have emerged, most are limited to single cognitive tasks, lack multi-dimensional data fusion analysis, and cannot provide comprehensive, objective, and personalized assessment results. Summary of the Invention

[0004] This application provides a big data-based auxiliary assessment method and system for cognitive impairment, which can significantly improve the accuracy and sensitivity of early screening for cognitive impairment while achieving automation and objectivity in the assessment process, and provides interpretable assessment results and personalized intervention recommendations.

[0005] To achieve the above objectives, this application adopts the following technical solution: Firstly, a big data-based auxiliary assessment method for cognitive impairment is provided. The method includes: acquiring dynamic data of the subject to be assessed, whereby the dynamic data represents the subject's basic information and cognitive information; determining the dynamic deviation from a healthy baseline model based on the dynamic data; performing a cognitive impairment assessment calculation based on the dynamic deviation to obtain the subject's cognitive impairment risk level; identifying the damaged area based on the dynamic data and the dynamic deviation, whereby the damaged area is the region of the subject's brain cognitive impairment function that is impaired; and determining intervention recommendations based on the subject's cognitive impairment risk level and the damaged area.

[0006] Preferably, determining the dynamic deviation from the health baseline model based on dynamic data includes: constructing a health baseline model based on the historical dynamic data of the subject to be evaluated, the health baseline model being used to characterize its multi-dimensional data distribution characteristics under a cognitively stable state; extracting multi-dimensional temporal features from the dynamic data, the multi-dimensional temporal features including the trend of cognitive function changes, behavioral rhythm fluctuation characteristics, and the coefficient of variation of physiological indicators; calculating the deviation measure of the multi-dimensional temporal features relative to the health baseline model; and weighting and fusing the deviation measures to generate a comprehensive dynamic deviation.

[0007] Preferably, the calculation of deviation measures of multi-dimensional time-series features relative to the healthy baseline model includes: Calculate the Mahalanobis distance of the multidimensional temporal features to the baseline in the healthy baseline model as a deviation measure; or, When the healthy baseline model is a probability density model, the negative logarithm of the probability density value of the current multidimensional time series features under the healthy baseline model is calculated as a deviation measure.

[0008] Preferably, the deviation metrics are weighted and fused to generate a comprehensive dynamic deviation, including: The dynamic weight of each deviation metric is determined based on the significance coefficient of the anomaly represented by each deviation metric. The overall dynamic deviation is calculated by weighting all deviation measures based on their dynamic weights.

[0009] Preferably, cognitive impairment risk level of the subject to be assessed is obtained by performing a cognitive impairment assessment based on dynamic deviation, including: The comprehensive dynamic deviation and multi-dimensional temporal features are input together into the pre-trained deep neural network for cognitive impairment risk assessment. The deep neural network outputs the cognitive impairment risk level of the subject to be evaluated, which includes low risk, medium risk, and high risk.

[0010] Preferably, the damaged area is determined based on dynamic data and dynamic deviation, including: The dynamic weights of the deviation metric are compared with a first weight threshold, and the first feature is determined based on the comparison result. The first feature is mapped to a predefined cognitive functional domain.

[0011] Preferably, the dynamic weights of the deviation metric are compared with a first weight threshold, and a first feature is determined based on the comparison result, including: If the dynamic weight of the deviation metric is greater than or equal to the first weight threshold, the multi-dimensional temporal feature corresponding to the deviation metric is determined as the first feature.

[0012] Preferably, the cognitive functional domains include at least one of the following: memory, attention, executive function, language ability, and visuospatial ability.

[0013] Preferably, intervention recommendations are determined based on the cognitive impairment risk level and affected areas of the subject to be assessed, including: Based on the risk level and affected area of ​​the object to be evaluated, a path search is performed in the knowledge graph to match and obtain a combination of intervention measures. The nodes of the knowledge graph include risk level, affected area, type of intervention measure and intervention activity, and the edges represent the relationship and strength of the relationship between nodes. Adjust the intensity and frequency of intervention measures according to the risk level, and generate intervention recommendations.

[0014] Based on the aforementioned technologies, early and accurate screening of cognitive impairment can be achieved by constructing individualized health baselines. Using multi-dimensional dynamic data, risk levels are quantified and affected areas are located, making the assessment results objective and interpretable. Ultimately, personalized intervention recommendations are generated, forming an assessment-intervention closed loop and effectively improving the level of cognitive health management.

[0015] Secondly, a big data-based auxiliary assessment system for cognitive impairment is provided, comprising: Data acquisition module: used to acquire dynamic data of the object to be evaluated. The dynamic data is used to represent the basic information and cognitive information of the object to be evaluated. Deviation determination module: Used to determine the dynamic deviation from the health baseline model based on dynamic data; Assessment and calculation module: used to perform cognitive impairment assessment calculations based on dynamic deviation to obtain the cognitive impairment risk level of the subject to be assessed; Region identification module: used to identify the damaged region based on dynamic data and dynamic deviation. The damaged region is the area of ​​the brain cognitive impairment function of the subject to be evaluated. Suggested output module: Used to determine intervention recommendations based on the cognitive impairment risk level and affected areas of the subject to be assessed.

[0016] The solution provided in the second aspect above is used to implement the method provided in the first aspect above, and its specific implementation will not be described in detail here. The technical effects corresponding to any implementation method of the solution provided in the second aspect above can be found in the technical effects corresponding to any implementation method of the first aspect above, and will not be described in detail here.

[0017] It should be noted that any of the possible implementations of any of the above aspects can be combined, provided that the solutions do not contradict each other. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application, 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 this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A logical diagram illustrating a big data-based auxiliary assessment method for cognitive impairment provided in this application embodiment; Figure 2 A flowchart illustrating a big data-based auxiliary assessment method for cognitive impairment provided in this application embodiment; Figure 3 This is a schematic diagram of the structure of a big data-based auxiliary assessment system for cognitive impairment provided in an embodiment of this application. Detailed Implementation

[0020] In the embodiments of this application, in order to clearly describe the technical solutions of the embodiments of this application, the terms "first" and "second" are used to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different. The technical features described by "first" and "second" have no sequential or size order.

[0021] In the embodiments of this application, the words "exemplarily" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplarily" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of the words "exemplarily" or "for example" is intended to present the relevant concepts in a specific manner to facilitate understanding.

[0022] In the embodiments of this application, at least one can also be described as one or more, and multiple can be two, three, four or more, and this application does not impose any restrictions.

[0023] Furthermore, the network architecture and scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of network architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0024] The solutions provided by the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0025] Figure 1This is a logical diagram illustrating an auxiliary assessment method for cognitive impairment based on big data, provided in an embodiment of this application.

[0026] like Figure 1 As shown, the entire assessment process starts with the dynamic data of the subject to be assessed, and through analysis and calculation, finally outputs the results of the cognitive impairment assessment.

[0027] Optionally, before outputting the results of the cognitive impairment assessment, the method may also include a data visualization step to generate a trend graph of the user's cognitive function over time and compare the current assessment results with historical assessment results to more intuitively show the evolution of their cognitive state.

[0028] Specifically, the analysis and calculation process includes: first, cleaning and standardizing the dynamic data; then, using a deep learning model to automatically extract multidimensional features related to cognitive function from the processed data; and finally, inputting these features into a pre-trained cognitive impairment risk assessment model, and obtaining the final assessment result through model inference.

[0029] For example, for an elderly user living alone, their dynamic data may include: sleep duration and heart rate variability data collected via a smartwatch, daily activity range and social call frequency recorded via a mobile app, and scores from a weekly online memory test. This data collectively constitutes the user's dynamic dataset and is input into the assessment process.

[0030] like Figure 2 As shown in the embodiments of this application, a big data-based auxiliary assessment method for cognitive impairment may include: S201: Obtain dynamic data of the object to be evaluated.

[0031] Among them, dynamic data is used to represent the basic information and cognitive information of the object to be evaluated.

[0032] Specifically, the dynamic data is acquired through various data collection channels, including but not limited to wearable devices, mobile terminals, smart home devices, electronic medical record systems, and online cognitive assessment platforms.

[0033] Optionally, before acquiring dynamic data, the method further includes sending a data collection authorization request to the subject to be evaluated or their guardian, and initiating the data collection process after obtaining explicit authorization; all collected dynamic data is encrypted and anonymized during transmission and storage to protect user privacy.

[0034] In some embodiments, the dynamic data collection is a continuous longitudinal collection aimed at constructing a personal health profile of the subject to be assessed over a preset time period. In this way, the system can capture subtle changes and long-term decline trends in cognitive function, rather than relying solely on static snapshots of a single assessment, thereby significantly improving the sensitivity and accuracy of early screening. Optionally, the subject to be assessed refers to individuals at risk of cognitive impairment or requiring cognitive function screening, such as middle-aged and elderly individuals, those with a family history of neurological diseases, or those experiencing subjective cognitive decline.

[0035] Basic information refers to background data related to cognitive health, including demographic characteristics such as age, gender, education level, medical history, medication history, family history, and lifestyle information.

[0036] Cognitive information refers to data that directly or indirectly reflects the state of cognitive function, including cognitive domain scores obtained through standardized cognitive scale assessments, daily behavioral trajectory data such as activity patterns and frequency of social interactions, and physiological indicators related to cognitive function such as sleep structure and EEG signal characteristics.

[0037] S202: Based on dynamic data, determine the dynamic deviation from the health baseline model.

[0038] In some embodiments, a health baseline model is constructed based on the historical dynamic data of the object to be evaluated. The health baseline model is used to characterize its multi-dimensional data distribution characteristics under a cognitively stable state. Multi-dimensional time-series features are extracted from the dynamic data. The multi-dimensional time-series features include the trend of cognitive function change, the characteristics of behavioral rhythm fluctuations, and the coefficient of variation of physiological indicators. The deviation measure of the multi-dimensional time-series features relative to the health baseline model is calculated. The deviation measures are weighted and fused to generate a comprehensive dynamic deviation.

[0039] A health baseline model is a personalized mathematical model built on the historical data of the subject being assessed, used to characterize the inherent patterns and fluctuation range of multimodal data in a state of stable or healthy cognitive function.

[0040] Specifically, the health baseline model is constructed using unsupervised learning algorithms, such as Gaussian mixture models or variational autoencoders, to train the dynamic data of the object to be evaluated within a preset historical period, learn and fit the joint probability distribution of its multi-dimensional temporal features, and the model obtained after training is the health baseline model.

[0041] Multidimensional data distribution characteristics refer to the statistical properties of multiple time-series features of different dimensions in a steady state, including but not limited to the mean and variance of each feature, as well as the covariance or correlation structure between different features.

[0042] The coefficient of variation of a physiological indicator is a relative statistic that measures the dispersion of physiological indicator data. It is calculated as the ratio of the standard deviation of the physiological indicator data to its mean, and is used to assess the stability of the indicator's fluctuations over a period of time.

[0043] For example, the Mahalanobis distance of multidimensional temporal features to the baseline in the healthy baseline model is calculated as a deviation measure; Alternatively, if the health baseline model is a probability density model, calculate the negative logarithm of the probability density value of the current multidimensional time series features under the health baseline model, and use it as a deviation measure.

[0044] Deviation metric refers to a numerical indicator used to quantify the degree of deviation of a single multidimensional time-series feature or a combination thereof from the normal state defined by a healthy baseline model.

[0045] Specifically, the currently collected multi-dimensional time-series features are taken as a data point and input into the constructed health baseline model. If the baseline model is a Gaussian distribution model, the Mahalanobis distance of the data point is calculated using the mean vector and covariance matrix of the model. If the baseline model is another probability density model, the data point is substituted into the model function to calculate its probability density value.

[0046] For example, the dynamic weight of each deviation metric is determined based on the significance coefficient of the anomaly represented by each deviation metric; The overall dynamic deviation is calculated by weighting all deviation measures based on their dynamic weights.

[0047] Specifically, the significance coefficient corresponding to each deviation metric is input into a pre-trained attention network. This network, through non-linear transformation and the Softmax activation function, outputs a normalized weight value for each deviation metric, which is the dynamic weight. Here, dynamic weight refers to weights that are not fixed but are calculated in real-time based on the degree of anomaly of each feature in the current dynamic data, adaptively highlighting the most critical risk indicators.

[0048] The significance coefficient is the raw numerical value used to quantify the degree of anomaly represented by a single deviation measure. It can usually be the deviation measure itself (such as Mahalanobis distance or negative log probability) or the value after normalization. The larger the value, the more significant the anomaly.

[0049] Specifically, the dynamic weights for each deviation metric are determined by inputting the significance coefficients corresponding to each deviation metric into a pre-trained attention network. This network outputs the normalized weight values ​​for each deviation metric through nonlinear transformation and the Softmax activation function, which are the dynamic weights.

[0050] S203: Perform cognitive impairment assessment calculations based on dynamic deviation to obtain the cognitive impairment risk level of the subject to be assessed.

[0051] In some embodiments, the integrated dynamic deviation and multi-dimensional temporal features are input together into a pre-trained deep neural network for cognitive impairment risk assessment. The deep neural network outputs the cognitive impairment risk level of the subject to be assessed, which includes low risk, medium risk, and high risk.

[0052] Specifically, the comprehensive dynamic deviation is concatenated or fused with multi-dimensional temporal features to form a unified input vector. This input vector is then fed into a pre-trained deep neural network for cognitive impairment risk assessment for forward propagation calculation. Through its multi-layer nonlinear transformations, the network ultimately outputs a probability distribution for each risk level.

[0053] A pre-trained deep neural network for cognitive impairment risk assessment refers to a deep learning model trained using large-scale, labeled dynamic data from both cognitively and non-cognitively impaired populations. Its internal parameters have been optimized to convergence. The input consists of the comprehensive dynamic deviation and multi-dimensional temporal features used to calculate this deviation. The output is a vector containing the probability values ​​of the assessed object belonging to three categories: low-risk, medium-risk, and high-risk.

[0054] Specifically, the probability distribution vector output by the network is analyzed, and the risk level with the highest probability value is selected as the final evaluation result. Simultaneously, this highest probability value is used as the confidence score for this evaluation and output along with the data. For example, if the output vector is [low risk: 0.1, medium risk: 0.3, high risk: 0.6], then the final determined risk level is "high risk," with a confidence score of 0.6.

[0055] The cognitive impairment risk level of an individual being assessed is a quantitative assessment result that characterizes the likelihood of the individual currently having cognitive impairment or developing cognitive impairment within a specific future timeframe. This level serves as a classification label to guide subsequent intervention strategies, and its determination is based on a comprehensive analysis of multi-dimensional data using deep neural networks.

[0056] S204: Determine the damaged area based on dynamic data and dynamic deviation.

[0057] The damaged area refers to the area of ​​the brain where cognitive impairment is impaired in the subject being evaluated.

[0058] Specifically, by analyzing the contribution of each feature in calculating dynamic deviation, we can trace back to the key factors that lead to abnormal cognitive states.

[0059] By using the dynamic weights of a defined deviation metric, one or more key features that contribute the most to the overall deviation are identified, and these key features are matched with preset cognitive functional domains to locate the impaired cognitive functional areas.

[0060] The damaged area refers to a set of one or more core cognitive functional domains in which the cognitive function of the subject under assessment shows significant decline or abnormality. This result provides a direct target for subsequent personalized intervention.

[0061] In some embodiments, the dynamic weight of the deviation metric is compared with a first weight threshold, a first feature is determined based on the comparison result, and the first feature is mapped to a preset cognitive functional domain.

[0062] The first feature refers to the dynamic weight of its corresponding deviation metric reaching or exceeding a preset threshold, thus being judged as a key multi-dimensional temporal feature that significantly contributes to the current cognitive state abnormality.

[0063] The first weight threshold is a preset value used to filter out features that have a decisive impact on the assessment results of cognitive impairment. This threshold can be set according to the sensitivity requirements of the model, for example, to 0.2, or dynamically determined according to the distribution of all dynamic weights, such as selecting the top 20% of features with the highest weight values.

[0064] A pre-defined cognitive functional domain refers to a pre-established knowledge base or mapping table that associates multi-dimensional temporal features with known cognitive functional categories. This knowledge base is constructed based on neuroscience and clinical medicine knowledge.

[0065] For example, the "declining trend of memory test scores" can be mapped to the memory domain, and the "declining regularity of daily activities" can be mapped to the executive function domain.

[0066] Specifically, the system iterates through all dynamic weights that deviate from the metric and compares each dynamic weight with a first weight threshold.

[0067] Specifically, the first feature is mapped to a preset cognitive functional domain by querying the knowledge base or mapping table corresponding to the preset cognitive functional domain, to find the cognitive functional domain associated with each first feature, and to identify the cognitive functional domain as the damaged area.

[0068] Alternatively, in addition to filtering based on dynamic weights, all multi-dimensional temporal features and their dynamic deviations can be input into a pre-trained cognitive functional domain classification model, which directly outputs the damage probability of each cognitive functional domain, thereby determining the damaged area.

[0069] For example, if the dynamic weight of the deviation metric is greater than or equal to the first weight threshold, the multi-dimensional temporal feature corresponding to the deviation metric is determined as the first feature.

[0070] The cognitive functional domains include at least one of the following: memory, attention, executive function, language ability, and visuospatial ability.

[0071] Specifically, after mapping and filtering, the final output is a list of one or more cognitive functional domains that have been identified as impaired, which serves as the result of the impaired areas in this assessment.

[0072] S205: Based on the cognitive impairment risk level and affected areas of the subject to be assessed, determine intervention recommendations.

[0073] Specifically, the cognitive impairment risk level and the identified damaged area are used as input conditions. Through an intelligent decision support system, a set of personalized and executable cognitive intervention plans are generated for the individuals to be assessed.

[0074] An intervention recommendation is a comprehensive guidance plan that includes specific intervention activities, frequency and intensity of implementation, and expected goals. This plan is highly personalized, and its content is dynamically adjusted according to the risk level and impaired cognitive functional domains of the individual being assessed, aiming to delay or improve cognitive decline through non-pharmacological means.

[0075] In some embodiments, based on the risk level and affected area of ​​the object to be evaluated, a path search is performed in the knowledge graph to match and obtain a combination of intervention measures. The nodes of the knowledge graph include risk level, affected area, type of intervention measure and intervention activity, and the edges represent the relationship and strength of the relationship between nodes. The intensity and frequency of intervention measures are adjusted according to the risk level to generate intervention recommendations.

[0076] Specifically, first, nodes corresponding to the "risk level" and "damaged area" of the object to be evaluated are located in the knowledge graph; then, starting from these two nodes, graph reasoning algorithms, such as path sorting algorithms or random walks, are used to explore nodes with strong associations to "intervention type" and "intervention activity"; finally, based on the association strength on the path, the searched intervention activities are sorted and combined to form a preliminary intervention combination.

[0077] A knowledge graph is a semantic network used to store and organize knowledge in the field of cognitive impairment intervention. It consists of entities (nodes) and relations (edges) and can express complex expert knowledge and clinical guidelines in a structured form, such as "what kind of intervention should be used for what kind of damaged area and what kind of risk level".

[0078] Specifically, the knowledge graph is constructed by extracting knowledge from multi-source heterogeneous data, including publicly available clinical practice guidelines, authoritative medical literature, experiential knowledge from neuroscience experts, and anonymized real-world intervention case data. Entities and relationships are identified from these data sources using natural language processing and information extraction techniques and then populated into the knowledge graph.

[0079] like Figure 3 The system shown is an auxiliary assessment system for cognitive impairment based on big data. The system includes: Data acquisition module 301: used to acquire dynamic data of the object to be evaluated. The dynamic data is used to represent the basic information and cognitive information of the object to be evaluated. Deviation determination module 302: used to determine the dynamic deviation from the health baseline model based on dynamic data; Assessment Calculation Module 303: Used to perform cognitive impairment assessment calculations based on dynamic deviation to obtain the cognitive impairment risk level of the subject to be assessed; Region determination module 304: used to determine the damaged region based on dynamic data and dynamic deviation. The damaged region is the area of ​​the brain cognitive impairment function of the subject to be evaluated. Suggested output module 305: Used to determine intervention recommendations based on the cognitive impairment risk level and affected areas of the subject to be assessed.

[0080] This application embodiment can divide the machine vision-based glass edge defect detection system into functional modules according to the above method embodiment. For example, each function can be divided into its own functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. It should be noted that the module division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0081] The method steps in this embodiment can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, portable hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary embodiment couples a storage medium to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. Additionally, the ASIC can reside in a network device. Alternatively, the processor and storage medium can exist as discrete components in the network device. In the above embodiments, implementation can be entirely or partially achieved through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented entirely or partially as a computer program product. A computer program product includes one or more computer programs or instructions. When a computer program or instruction is loaded and executed on a computer, all or part of the processes or functions of the embodiments of this application are performed. The computer may be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user equipment, or other programmable module. The computer program or instructions may be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, a computer program or instructions may be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium may be any available medium that a computer can access, or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; or an optical medium, such as a digital video disc (DVD); or a semiconductor medium, such as a solid-state drive (SSD). The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0082] Since the vehicle control device in the embodiments of the present invention can be applied to the above method, the technical effects it can achieve can also be referred to the above method embodiments, and the embodiments of the present invention will not be repeated here. The above are only specific embodiments of this application, but the protection scope of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims. The method steps in this embodiment can be implemented by hardware or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, which can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, portable hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor, thereby enabling the processor to read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. Alternatively, the ASIC can reside in a network device. Of course, the processor and storage medium can also exist as discrete components in the network device. In the above embodiments, implementation can be entirely or partially achieved through software, hardware, firmware, or any combination thereof. When implemented in software, it can be entirely or partially implemented in the form of a computer program product. A computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, all or part of the processes or functions of the embodiments of this application are performed. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user equipment, or other programmable modules. The computer program or instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, a computer program or instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media.The usable medium can be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; it can also be an optical medium, such as a digital video disc (DVD); or it can be a semiconductor medium, such as a solid-state drive (SSD). The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the scope of the technology disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A big data-based auxiliary assessment method for cognitive impairment, characterized in that, The method includes: Acquire dynamic data of the object to be evaluated, which is used to represent the basic information and cognitive information of the object to be evaluated; Based on the dynamic data, the dynamic deviation from the health baseline model is determined; Based on the dynamic deviation, a cognitive impairment assessment calculation is performed to obtain the cognitive impairment risk level of the subject to be assessed. Based on the dynamic data and dynamic deviation, the damaged area is determined, which is the area of ​​brain cognitive impairment function of the subject to be evaluated. Based on the cognitive impairment risk level and affected areas of the subjects to be assessed, intervention recommendations are determined.

2. The method according to claim 1, characterized in that, The determination of the dynamic deviation from the health baseline model based on the dynamic data includes: Based on the historical dynamic data of the object to be evaluated, the health baseline model is constructed. The health baseline model is used to characterize its multi-dimensional data distribution characteristics under a cognitively stable state. Multi-dimensional temporal features are extracted from the dynamic data, including trends in cognitive function, fluctuations in behavioral rhythms, and coefficients of variation of physiological indicators. Calculate the deviation measure of the multidimensional time-series features relative to the health baseline model; The deviation metrics are weighted and fused to generate a comprehensive dynamic deviation.

3. The method according to claim 2, characterized in that, The calculation of the deviation measure of the multi-dimensional time-series features relative to the health baseline model includes: The Mahalanobis distance of the multidimensional temporal features in the baseline of the health baseline model is calculated as the deviation measure. or, When the health baseline model is a probability density model, the negative logarithm of the probability density value of the current multidimensional time series feature under the health baseline model is calculated as the deviation measure.

4. The method according to claim 2, characterized in that, The weighted fusion of the deviation metrics to generate a comprehensive dynamic deviation includes: The dynamic weight of each deviation metric is determined based on the significance coefficient of the anomaly represented by each deviation metric. The overall dynamic deviation is obtained by weighting all deviation measures based on their dynamic weights.

5. The method according to claim 1, characterized in that, The cognitive impairment risk level of the subject under assessment is obtained by performing a cognitive impairment assessment based on the dynamic deviation, including: The comprehensive dynamic deviation and multi-dimensional temporal features are input together into the pre-trained deep neural network for cognitive impairment risk assessment. The deep neural network outputs the cognitive impairment risk level of the subject to be evaluated, and the risk level includes low risk, medium risk, and high risk.

6. The method according to claim 4, characterized in that, The process of determining the damaged area based on the dynamic data and dynamic deviation includes: The dynamic weight of the deviation metric is compared with a first weight threshold, and a first feature is determined based on the comparison result; the first feature is then mapped to a preset cognitive functional domain.

7. The method according to claim 6, characterized in that, The step of comparing the dynamic weight of the deviation metric with a first weight threshold and determining the first feature based on the comparison result includes: If the dynamic weight of the deviation metric is greater than or equal to the first weight threshold, the multi-dimensional temporal feature corresponding to the deviation metric is determined as the first feature.

8. The method according to claim 6, characterized in that, The cognitive functional domains include at least one of the following: memory, attention, executive function, language ability, and visuospatial ability.

9. The method according to claim 1, characterized in that, The intervention recommendations, based on the cognitive impairment risk level and affected areas of the subject to be assessed, include: Based on the risk level and affected area of ​​the object to be evaluated, a path search is performed in the knowledge graph to match and obtain a combination of intervention measures. The nodes of the knowledge graph include risk level, affected area, type of intervention measure and intervention activity, and the edges represent the association relationship and association strength between the nodes. Based on the risk level, adjust the intensity and frequency of the corresponding intervention measures to generate intervention recommendations.

10. A big data-based auxiliary assessment system for cognitive impairment, characterized in that, The system includes: Data acquisition module: used to acquire dynamic data of the object to be evaluated, the dynamic data being used to represent the basic information and cognitive information of the object to be evaluated; Deviation determination module: used to determine the dynamic deviation from the health baseline model based on the dynamic data; Assessment and calculation module: used to perform cognitive impairment assessment calculations based on the dynamic deviation to obtain the cognitive impairment risk level of the subject to be assessed; Region determination module: used to determine the damaged region based on the dynamic data and dynamic deviation, wherein the damaged region is the region of the brain cognitive impairment function of the subject to be evaluated; Suggested output module: Used to determine intervention recommendations based on the cognitive impairment risk level and affected areas of the subject to be assessed.

Images