Cognitive disorder detection device based on artificial intelligence, computer equipment and medium

Abstract

The invention belongs to the field of artificial intelligence and the field of digital medical treatment, and relates to a cognitive disorder detection device based on artificial intelligence, computer equipment and a storage medium. Preprocessing the original electroencephalogram signal to obtain an electroencephalogram signal sequence; adding classification marks and position codes to the electroencephalogram signal sequence to obtain a target electroencephalogram signal sequence; processing the target electroencephalogram signal sequence based on an anomaly detection layer in an anomaly detection model to obtain target features; performing prediction processing on the target features based on a full connection layer in the anomaly detection model to obtain a prediction result; generating a numerical comparison result of the prediction result and a preset threshold value; and generating a shallow cognitive impairment detection result corresponding to the user based on the numerical comparison result. In addition, shallow cognitive impairment detection results can be stored in the block chain. According to the method, the shallow cognitive impairment detection mode based on the anomaly detection model is adopted, so that the accuracy of shallow cognitive impairment detection is improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence development technology and the field of digital medicine, and in particular to an artificial intelligence-based cognitive impairment detection device, computer equipment and storage medium. Background Art

[0002] In the medical field, electroencephalogram (EEG) has been widely used in the detection of mild cognitive impairment (MCI) as a non-invasive neurophysiological monitoring method. Traditional EEG analysis methods mainly rely on time domain, frequency domain and time-frequency domain analysis, which provide an important basis for diagnosis by identifying the specific brain wave characteristics of MCI patients. However, with the in-depth research of technology and the growth of application demand, the existing technology has gradually revealed its limitations in the detection of mild cognitive impairment.

[0003] The traditional EEG analysis based on feature extraction for the detection of mild cognitive impairment (MCI) usually requires professionals to have deep knowledge of neuroelectrophysiology and signal processing skills, and to extract feature parameters such as amplitude and frequency through complex calculation processes. This highly manual method is not only inefficient, but also easily leads to subjectivity and instability in the test results, making it difficult to meet the clinical needs for fast and accurate test results. Summary of the invention

[0004] The purpose of the embodiments of the present application is to propose a cognitive impairment detection device, computer equipment and storage medium based on artificial intelligence, so as to solve the technical problem that the existing processing method for detecting mild cognitive impairment (MCI) based on feature extraction electroencephalogram analysis is highly dependent on manual labor, which is not only inefficient but also easily leads to subjectivity and instability of the detection results.

[0005] In order to solve the above technical problems, the present application embodiment provides a cognitive impairment detection device based on artificial intelligence, which adopts the following technical solutions:

[0006] The first acquisition module is used to acquire the user's original EEG signal;

[0007] A preprocessing module, used for preprocessing the original EEG signal to obtain a corresponding EEG signal sequence;

[0008] A first processing module is used to add classification tags and position codes to the EEG signal sequence to obtain a corresponding target EEG signal sequence;

[0009] A second processing module is used to process the target EEG signal sequence based on the anomaly detection layer in the trained anomaly detection model to obtain a target feature corresponding to the target EEG signal sequence;

[0010] A prediction module, used to perform prediction processing on the target feature based on the fully connected layer in the anomaly detection model to obtain a corresponding prediction result;

[0011] A first generating module, used to obtain a preset threshold value and generate a numerical comparison result between the prediction result and the preset threshold value;

[0012] The second generating module is used to generate a mild cognitive impairment detection result corresponding to the user based on the numerical comparison result.

[0013] Furthermore, the preprocessing module includes:

[0014] A filter submodule, used for filtering the original EEG signal to obtain a corresponding first EEG signal;

[0015] A segmentation submodule, used to perform channel segmentation processing on the first EEG signal to obtain a corresponding second EEG signal;

[0016] a flattening submodule, used for performing a data flattening operation on the second EEG signal to obtain a corresponding third EEG signal;

[0017] A projection submodule, used for performing linear projection processing on the third EEG signal to obtain a corresponding fourth EEG signal;

[0018] A determination submodule is used to use the fourth EEG signal as the target EEG signal.

[0019] Furthermore, the second generating module includes:

[0020] A first judgment submodule, used to judge whether the numerical comparison result is that the prediction result is greater than the preset threshold;

[0021] A first generating submodule, configured to generate a first mild cognitive impairment detection result indicating that the user has mild cognitive impairment if the predicted result is greater than the preset threshold;

[0022] The second generating submodule is configured to generate a second mild cognitive impairment detection result indicating that the user does not have mild cognitive impairment if the predicted result is less than the preset threshold.

[0023] Furthermore, the artificial intelligence-based cognitive impairment detection device also includes:

[0024] The second acquisition module is used to acquire pre-constructed EEG signal samples;

[0025] A division module, used for dividing the EEG signal samples into a training set and a test set;

[0026] A third acquisition module is used to acquire a preset initial anomaly detection model and initialize the initial anomaly detection model;

[0027] A third processing module, configured to input the training set into the initial anomaly detection model for forward propagation, and calculate an error between a model output of the initial anomaly detection model and a true label of the training set;

[0028] A fourth processing module, configured to calculate a gradient based on the error, update a model weight of the initial anomaly detection model through a back propagation algorithm, and optimize the error using a minimization strategy and a maximization strategy;

[0029] A fifth processing module is used to repeat the forward propagation and back propagation processes until a preset iteration condition is met to obtain a corresponding specified anomaly detection model;

[0030] A testing module, used for testing the specified anomaly detection model based on the test set;

[0031] A determination module is used to use the specified anomaly detection model as the anomaly detection model if the specified anomaly detection model passes the test.

[0032] Furthermore, the second acquisition module includes:

[0033] The first acquisition submodule is used to acquire the pre-collected initial EEG signal samples;

[0034] The calling submodule is used to call the preset sample to build the model;

[0035] A construction submodule, used for performing sample construction processing on the initial EEG signal sample based on the sample construction model to obtain a corresponding expanded sample;

[0036] The third generating submodule is used to generate the EEG signal sample based on the initial EEG signal sample and the expanded sample.

[0037] Furthermore, the test module includes:

[0038] A second acquisition submodule is used to input the test set into the specified anomaly detection model and obtain a performance indicator value of the specified anomaly detection model; wherein the performance indicator value is a value corresponding to a preset performance indicator, and the number of the performance indicators includes a plurality;

[0039] A second judgment submodule is used to judge whether all the performance indicator values ​​are within the corresponding preset value range;

[0040] A first determination submodule, configured to determine that the specified anomaly detection model passes the test if all values ​​are within a corresponding preset value range;

[0041] The second determination submodule is used to determine that the specified anomaly detection model fails the test if the values ​​are not all within the corresponding preset value range.

[0042] Furthermore, the artificial intelligence-based cognitive impairment detection device also includes:

[0043] A third generating module, configured to generate a corresponding target detection report based on the mild cognitive impairment detection result;

[0044] A fourth acquisition module, used to acquire the communication information of the user;

[0045] A push module is used to push the target detection report to the user based on the communication information.

[0046] Furthermore, the third generation module includes:

[0047] A conversion submodule, used for converting the shallow cognitive impairment detection result into a target detection result corresponding to a preset understanding style;

[0048] A third acquisition submodule is used to acquire medical advice information corresponding to the mild cognitive impairment test result;

[0049] A fourth acquisition submodule is used to acquire lifestyle recommendation information corresponding to the mild cognitive impairment test result;

[0050] The fifth acquisition submodule is used to obtain a preset test report template;

[0051] The filling submodule is used to fill the target detection result, the medical advice information and the lifestyle advice information into the detection report template to obtain the target detection report.

[0052] In order to solve the above technical problems, an embodiment of the present application also provides a computer device, including a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the functions of each module in the above-mentioned artificial intelligence-based cognitive impairment detection device when executing the computer-readable instructions.

[0053] In order to solve the above technical problems, an embodiment of the present application also provides a computer-readable storage medium, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor, the functions of each module in the above-mentioned artificial intelligence-based cognitive impairment detection device are realized.

[0054] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0055] In the present application, the cognitive impairment detection device based on artificial intelligence first obtains the original EEG signal of the user; and pre-processes the original EEG signal to obtain the corresponding EEG signal sequence; then adds classification tags and position codes to the EEG signal sequence to obtain the corresponding target EEG signal sequence; then processes the target EEG signal sequence based on the abnormal detection layer in the trained abnormal detection model to obtain the target feature corresponding to the target EEG signal sequence; subsequently, the target feature is predicted and processed based on the fully connected layer in the abnormal detection model to obtain the corresponding prediction result; further obtains the preset threshold, and generates a numerical comparison result between the prediction result and the preset threshold; finally, based on the numerical comparison result, generates a shallow cognitive impairment detection result corresponding to the user. The present application obtains the target EEG signal sequence by pre-processing the original EEG signal of the user obtained and adding classification tags and position codes, and then uses the trained abnormal detection model to predict the target EEG signal sequence to obtain the corresponding prediction result, and finally analyzes the prediction result based on the preset threshold, so as to quickly and accurately generate the shallow cognitive impairment detection result corresponding to the user. The present application adopts a method of detecting mild cognitive impairment based on an abnormality detection model, which no longer requires high reliance on manual processing and effectively improves the sensitivity and accuracy of mild cognitive impairment detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0057] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0058] Figure 2 is a structural schematic diagram of an embodiment of a cognitive impairment detection device based on artificial intelligence according to the present application;

[0059] Figure 3 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0061] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0062] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0063] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0064] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0065] Terminal devices 101, 102, 103 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 (Moving Picture Experts Group Audio Layer IV) players, laptop computers, desktop computers, etc.

[0066] The server 105 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal devices 101 , 102 , and 103 .

[0067] It should be noted that the artificial intelligence-based cognitive impairment detection device provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the artificial intelligence-based cognitive impairment detection device is generally set in the server / terminal device.

[0068] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0069] Continue to refer Figure 2 , showing a structural diagram of an embodiment of an artificial intelligence-based cognitive impairment detection device according to the present application. According to different needs, the order of the modules in the structural diagram can be changed, and some modules can be omitted. The artificial intelligence-based cognitive impairment detection device provided in the embodiment of the present application can be applied to any processing scenario that requires shallow cognitive impairment detection, and the artificial intelligence-based cognitive impairment detection device can be applied to products in these scenarios. The artificial intelligence-based cognitive impairment detection device 200 includes: a first acquisition module 201, a preprocessing module 202, a first processing module 203, a second processing module 204, a prediction module 205, a first generation module 206 and a second generation module 207. Among them:

[0070] The first acquisition module 201 is used to acquire the original EEG signal of the user.

[0071] In this embodiment, the artificial intelligence-based cognitive impairment detection device runs on an electronic device (e.g. Figure 1The server / terminal device shown in the figure) can obtain the original EEG signal through a wired connection or a wireless connection. It should be pointed out that the above-mentioned wireless connection method may include but is not limited to 3G / 4G / 5G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future. The present application can be applied to the field of digital medicine, and can be specifically used in medical business scenarios to assess whether a user is at risk of shallow cognitive impairment. The above-mentioned user may be a potential user who may be at risk of shallow cognitive impairment to be detected, and the above-mentioned original EEG signal may be the user's brain electrical activity during daily activities or specific tasks during the evaluation test of shallow cognitive impairment using an EEG (electroencephalogram) device, and the corresponding multi-channel EEG signal data obtained. Each channel represents the brain electrical activity recorded by electrodes at different positions on the scalp.

[0072] The preprocessing module 202 is used to preprocess the original EEG signal to obtain a corresponding EEG signal sequence.

[0073] In this embodiment, the specific implementation process of preprocessing the original EEG signal to obtain the corresponding EEG signal sequence will be further described in detail in subsequent specific embodiments of the present application and will not be elaborated on here.

[0074] The first processing module 203 is used to add classification tags and position codes to the EEG signal sequence to obtain a corresponding target EEG signal sequence.

[0075] In this embodiment, a CLS tag (classification tag) can be added at the beginning of each time window of the EEG signal sequence to represent the global information of the entire time window. This step helps to retain global context information in the subsequent Transformer model. Further, position encoding is added to each time window to retain time and channel information. This position encoding enables the Transformer model to utilize the temporal and spatial order information of the input data and enhance the learning ability of the model.

[0076] The second processing module 204 is used to process the target EEG signal sequence based on the anomaly detection layer in the trained anomaly detection model to obtain target features corresponding to the target EEG signal sequence.

[0077] In this embodiment, the target EEG signal sequence is input into a trained anomaly detection model. The anomaly detection model is specifically composed of an Anomaly Transformer and a fully connected layer, and the anomaly detection layer refers to the Anomaly Transformer. Anomaly Transformer is composed of multiple layers of Anomaly-Attention blocks and feedforward neural network layers stacked alternately. Each layer calculates the Attention weight of the input signal and further processes it through a feedforward neural network. The stacking of these layers enables the model to extract and integrate the deep features of the signal layer by layer, that is, the above-mentioned target features.

[0078] The prediction module 205 is used to perform prediction processing on the target feature based on the fully connected layer in the anomaly detection model to obtain a corresponding prediction result.

[0079] In this embodiment, the output of the anomaly detection layer, i.e., the target feature, is input into the fully connected layer, and the corresponding prediction result is output through the fully connected layer. The prediction result output by the fully connected layer is a probability distribution, which represents the prediction probability for each category ("there is mild cognitive impairment" and "there is no mild cognitive impairment").

[0080] The first generating module 206 is used to obtain a preset threshold and generate a numerical comparison result between the prediction result and the preset threshold.

[0081] In this embodiment, there is no specific limitation on the value of the preset threshold, which can be set to 0.5, and the threshold can be adjusted according to the actual situation to optimize the performance of the model. For example, the optimal threshold can be determined by cross-validation or other methods. The numerical comparison result includes that the prediction result is greater than the preset threshold, or the prediction result is less than the preset threshold.

[0082] The second generating module 207 is used to generate a mild cognitive impairment detection result corresponding to the user based on the numerical comparison result.

[0083] In this embodiment, the specific implementation process of generating the mild cognitive impairment detection result corresponding to the user based on the numerical comparison result will be described in further detail in subsequent specific embodiments of the present application, and will not be elaborated on here.

[0084] In the present application, the cognitive impairment detection device based on artificial intelligence first obtains the original EEG signal of the user; and pre-processes the original EEG signal to obtain the corresponding EEG signal sequence; then adds classification tags and position codes to the EEG signal sequence to obtain the corresponding target EEG signal sequence; then processes the target EEG signal sequence based on the abnormal detection layer in the trained abnormal detection model to obtain the target feature corresponding to the target EEG signal sequence; subsequently, the target feature is predicted and processed based on the fully connected layer in the abnormal detection model to obtain the corresponding prediction result; further obtains the preset threshold, and generates a numerical comparison result between the prediction result and the preset threshold; finally, based on the numerical comparison result, generates a shallow cognitive impairment detection result corresponding to the user. The present application obtains the target EEG signal sequence by pre-processing the original EEG signal of the user obtained and adding classification tags and position codes, and then uses the trained abnormal detection model to predict the target EEG signal sequence to obtain the corresponding prediction result, and finally analyzes the prediction result based on the preset threshold, so as to quickly and accurately generate the shallow cognitive impairment detection result corresponding to the user. The present application adopts a method of detecting mild cognitive impairment based on an abnormality detection model, which no longer requires high reliance on manual processing and effectively improves the sensitivity and accuracy of mild cognitive impairment detection.

[0085] In some optional implementations, the preprocessing module 202 includes:

[0086] The filter submodule is used to filter the original EEG signal to obtain a corresponding first EEG signal.

[0087] In this embodiment, the original EEG signal can be filtered by using a filter to extract the signal components of a specific frequency band, thereby obtaining the first EEG signal. The specific frequency band can be determined according to actual business needs, such as alpha waves (8-13Hz), beta waves (13-30Hz), etc. The first EEG signal after filtering will remove most of the noise and retain the desired signal components of interest.

[0088] The segmentation submodule is used to perform channel segmentation processing on the first electroencephalogram signal to obtain a corresponding second electroencephalogram signal.

[0089] In this embodiment, the first EEG signal can be segmented into channels according to a preset time window. Each time window contains data frames within a certain time, and each frame contains data of multiple channels. Among them, the appropriate time window size can be determined according to the experimental requirements and analysis purposes. The size of the time window contains enough data frames for subsequent analysis while avoiding data redundancy.

[0090] The flattening submodule is used to perform a data flattening operation on the second EEG signal to obtain a corresponding third EEG signal.

[0091] In this embodiment, the data flattening operation refers to performing a flattening operation on the signal data of each channel, that is, converting a multi-dimensional signal into a one-dimensional sequence.

[0092] The projection submodule is used to perform linear projection processing on the third EEG signal to obtain a corresponding fourth EEG signal.

[0093] In this embodiment, linear projection techniques, such as PCA, LDA and other dimensionality reduction methods, can be used to remove redundant information in the third EEG signal while retaining the most important features. Linear projection processing can improve the quality of data and reduce the impact of noise and redundancy on subsequent analysis.

[0094] A determination submodule is used to use the fourth EEG signal as the target EEG signal.

[0095] The present application obtains the corresponding first EEG signal by filtering the original EEG signal; then performs channel segmentation processing on the first EEG signal to obtain the corresponding second EEG signal; then performs data flattening operation on the second EEG signal to obtain the corresponding third EEG signal; subsequently performs linear projection processing on the third EEG signal to obtain the corresponding fourth EEG signal, and uses the fourth EEG signal as the target EEG signal. The present application designs a specific preprocessing process based on the multi-channel characteristics of EEG signals. By filtering, channel segmentation, data flattening and linear projection processing on the original EEG signal, the spatial and temporal characteristics of the EEG signal are retained, and the representation quality of the EEG signal data is improved, so that the global and local features related to the detection of shallow cognitive impairment can be better captured, which helps to improve the accuracy of the detection of shallow cognitive impairment.

[0096] In some optional implementations, the second generating module 307 includes:

[0097] The first judgment submodule is used to judge whether the numerical comparison result is that the prediction result is greater than the preset threshold.

[0098] In this embodiment, the content of the above numerical comparison result includes that the prediction result is greater than the preset threshold, or the prediction result is less than the preset threshold.

[0099] The first generating submodule is used to generate a first mild cognitive impairment detection result indicating that the user has mild cognitive impairment if the predicted result is greater than the preset threshold.

[0100] In this embodiment, if the numerical comparison result is that the predicted result is greater than the preset threshold, the predicted result is converted into a first mild cognitive impairment detection result indicating that the user has mild cognitive impairment.

[0101] The second generating submodule is configured to generate a second mild cognitive impairment detection result indicating that the user does not have mild cognitive impairment if the predicted result is less than the preset threshold.

[0102] In this embodiment, if the numerical comparison result is that the predicted result is less than the preset threshold, the predicted result is converted into a first mild cognitive impairment detection result indicating that the user does not have mild cognitive impairment.

[0103] The present application determines whether the numerical comparison result is that the predicted result is greater than the preset threshold; if the predicted result is greater than the preset threshold, a first shallow cognitive impairment detection result is generated indicating that the user has shallow cognitive impairment; and if the predicted result is less than the preset threshold, a second shallow cognitive impairment detection result is generated indicating that the user does not have shallow cognitive impairment. The present application analyzes the content of the numerical comparison result, thereby quickly and accurately generating a shallow cognitive impairment detection result corresponding to the user, thereby ensuring the accuracy of the generated shallow cognitive impairment detection result.

[0104] In some optional implementations of this embodiment, the above-mentioned cognitive impairment detection device based on artificial intelligence further includes:

[0105] The second acquisition module is used to acquire pre-constructed EEG signal samples.

[0106] In this embodiment, the specific implementation process of obtaining the pre-constructed EEG signal samples will be described in further detail in subsequent specific embodiments of the present application, and will not be elaborated on here.

[0107] The division module is used to divide the EEG signal samples into a training set and a test set.

[0108] In this embodiment, the EEG signal samples can be divided into a training set and a test set according to a preset division ratio. The value of the division ratio is not specifically limited and can be set according to actual usage requirements, for example, it can be set to 7:3.

[0109] The third acquisition module is used to acquire a preset initial anomaly detection model and initialize the initial anomaly detection model.

[0110] In this embodiment, the above-mentioned initial anomaly detection model is specifically composed of an Anomaly Transformer model and a fully connected layer. Specifically, the overall structure of the above-mentioned Anomaly Transformer model is composed of alternately stacked Anomaly-Attention blocks and feedforward layers. This stacking structure is conducive to the model learning potential temporal associations from multi-level features. The Anomaly-Attention mechanism is a key component of the model, which considers both prior associations and sequence associations. Specifically, the above-mentioned Anomaly-Attention blocks include: Multi-layer stacking: The Anomaly Transformer model uses multi-layer stacked Anomaly-Attention blocks to process the input signal layer by layer. Each layer contains a self-attention mechanism and a feedforward neural network to capture contextual information and features in the input sequence. Attention weight calculation: In each layer, the Anomaly-Attention block evaluates the correlation between each element in the input sequence (i.e., the EEG signal at each time point) and other elements by calculating the Attention weights between them. These weights reflect the degree of mutual influence between signals at different time points. Feedforward neural network: The calculated Attention weights are used for weighted summation to generate a contextual representation of each element. These representations are then further processed by a feedforward neural network to extract higher-level features. Anomaly Attention Mechanism: The Anomaly Transformer model introduces an anomaly attention mechanism to distinguish normal and anomaly points through the distribution characteristics of the Attention weights. Anomalies usually show different patterns or distributions in the Attention weights than normal points.

[0111] The third processing module is used to input the training set into the initial anomaly detection model for forward propagation, and calculate the error between the model output of the initial anomaly detection model and the true label of the training set; wherein the error includes a reconstruction error and an associated difference.

[0112] In this embodiment, specifically, the process of inputting the training set into the initial anomaly detection model for forward propagation and calculating the error between the model output of the initial anomaly detection model and the true label of the training set includes: 1) calculating the prior association matrix. Gaussian kernel function: The Gaussian kernel function is used to calculate the prior relationship between time points. The Gaussian kernel function can evaluate the similarity or correlation between time points based on their relative distance. By adjusting the parameters of the Gaussian kernel (such as the standard deviation σ), the characteristics and patterns of different time series can be adapted. Calculation method: For each time point i, the prior association weight p_ij between it and other time points j in the sequence is calculated. These weights constitute a prior association matrix P, in which each element p_ij represents the prior association strength between time points i and j. 2) Calculating the sequence association matrix. Attention weight capture: The attention weights calculated by the Anomaly-Attention block capture the patterns and anomalies in the signal. These weights reflect the actual correlation between signals at different time points and are a representation of the model's intrinsic structure of the input sequence. 3) Generate a sequence association matrix: Normalize the Attention weights (such as using the Softmax function) to generate a sequence association matrix S. Each element s_ij in the matrix S represents the actual association strength between time points i and j. 4) Association difference evaluation. Calculate the difference (error): Use the symmetric KL divergence to calculate the difference between the prior association matrix P and the sequence association matrix S. This difference reflects the degree of deviation between the actual association pattern and the prior association pattern in the time series, and is an important indicator for evaluating the degree of abnormality in the signal. Abnormality degree evaluation: Based on the calculated association difference value, the degree of abnormality in the signal can be evaluated. Abnormal points usually result in larger association difference values ​​because their association patterns are significantly different from those of normal points.

[0113] The fourth processing module is used to calculate the gradient based on the error, update the model weight of the initial anomaly detection model through a back propagation algorithm, and optimize the error by adopting a minimization strategy and a maximization strategy.

[0114] In this embodiment, the process of back propagation includes: 1. Calculating gradients: Based on the errors (including reconstruction errors and associated differences) calculated by forward propagation, the gradients of the parameters in the model are calculated using the chain rule. 2. Updating model weights: Using the gradient descent algorithm or the Adam algorithm to update the weights and biases of the model, the reconstruction error is minimized by the minimization strategy and the associated difference is maximized by the maximization strategy, so that the model can better detect abnormal signals, thereby improving the detection ability of the model.

[0115] The fifth processing module is used to repeat the forward propagation and back propagation processes until the preset iteration conditions are met to obtain the corresponding specified anomaly detection model.

[0116] In this embodiment, the above-mentioned preset iteration condition may specifically refer to a preset number of iterations or satisfying a certain stopping condition (such as the performance on the validation set no longer significantly improves). There is no specific limitation on the selection of the above-mentioned number of iterations, which can be determined according to actual business needs.

[0117] A testing module is used to test the specified anomaly detection model based on the test set.

[0118] In this embodiment, the specific implementation process of testing the specified anomaly detection model based on the test set will be further described in detail in subsequent specific embodiments of the present application and will not be elaborated on here.

[0119] A determination module is used to use the specified anomaly detection model as the anomaly detection model if the specified anomaly detection model passes the test.

[0120] In this embodiment, if the specified anomaly detection model passes the test, it indicates that the model effect of the generated specified anomaly detection model has achieved the expected effect, and then the specified anomaly detection model will be used as the anomaly detection model. Among them, the present application introduces an Attention mechanism that combines prior association and serial association in the Anomaly Transformer model. Prior association represents the prior relationship between time points through a Gaussian kernel function, and serial association captures patterns and anomalies in the signal through Attention weights. This mechanism enables the model to more accurately identify subtle changes and abnormal patterns in EEG signals, which may be related to early symptoms of mild cognitive impairment (MCI). By capturing these subtle changes, the model can perform more effective detection in the early stages of MCI.

[0121] The present application obtains pre-constructed EEG signal samples and divides the EEG signal samples into a training set and a test set; then obtains a preset initial anomaly detection model and initializes the initial anomaly detection model; then inputs the training set into the initial anomaly detection model for forward propagation, and calculates the error between the model output of the initial anomaly detection model and the true label of the training set; subsequently, based on the error, the gradient is calculated, the model weight of the initial anomaly detection model is updated by the back propagation algorithm, and the error is optimized by minimization and maximization strategies; the forward propagation and back propagation processes are repeated until the preset iteration conditions are met to obtain the corresponding designated anomaly detection model; finally, the designated anomaly detection model is tested based on the test set; if the designated anomaly detection model passes the test, the designated anomaly detection model is used as the anomaly detection model. The present application obtains pre-constructed EEG signal samples, and then uses the Attention mechanism combining prior and sequence association and the optimized training strategy to train the initial anomaly detection model with the EEG signal samples to obtain the trained anomaly detection model, which effectively improves the detection performance and reliability of the anomaly detection model for the detection of shallow cognitive impairment.

[0122] In some optional implementations, the second acquisition module includes:

[0123] The first acquisition submodule is used to acquire the pre-collected initial EEG signal samples.

[0124] In this embodiment, the initial EEG signal sample may be sample data obtained by selecting 1.5s of EEG signal data from 500Hz EEG signal sampling data and manually annotating the data. The manually annotated labels include the presence of mild cognitive impairment or the absence of mild cognitive impairment.

[0125] The calling submodule is used to call the preset sample to build the model.

[0126] In this embodiment, the above-mentioned sample construction model can specifically adopt a model based on a generative adversarial network. The sample construction model is composed of a generator and a discriminator. The generator is responsible for generating fake samples similar to the initial EEG signal samples. The generator is a neural network, the input is random noise, and the output is the generated fake sample. The discriminator is responsible for distinguishing whether the input sample is real or a fake sample generated by the generator. The discriminator is also a neural network, the input is a sample, and the output is the probability that the sample is a real sample. The training process of the sample generation model includes: alternating training: in each round of training, the parameters of the generator are first fixed, and the discriminator is trained to distinguish between real samples and fake samples; then the parameters of the discriminator are fixed, and the generator is trained to generate more realistic fake samples, making it difficult for the discriminator to distinguish. Optimization goal: The goal of the generator is to generate fake samples that are as realistic as possible, so that the discriminator cannot distinguish; the goal of the discriminator is to distinguish between real samples and fake samples as accurately as possible. After completing the training of the sample construction model, a large number of fake samples can be generated by using the sample construction model. These fake samples are similar to real samples but have diversity, so that they can be used to enhance the diversity of sample data.

[0127] The construction submodule is used to perform sample construction processing on the initial EEG signal samples based on the sample construction model to obtain corresponding expanded samples.

[0128] In this embodiment, the initial EEG signal samples are processed by sample construction using a sample construction model to generate a large number of false samples (i.e., the above-mentioned expanded samples). These false samples are similar to real samples (i.e., the initial EEG signal samples) but have diversity, thereby effectively enhancing the diversity of sample data.

[0129] The third generating submodule is used to generate the EEG signal sample based on the initial EEG signal sample and the expanded sample.

[0130] In this embodiment, the initial EEG signal sample and the expanded sample may be integrated to obtain a corresponding integrated sample, and the integrated sample may be used as a final EEG signal sample.

[0131] The present application obtains the initial EEG signal samples collected in advance; then calls the preset sample construction model; then performs sample construction processing on the initial EEG signal samples based on the sample construction model to obtain corresponding expanded samples; and subsequently generates the EEG signal samples based on the initial EEG signal samples and the expanded samples. The present application can realize the intelligent and rapid generation of corresponding expanded samples based on the initial EEG signals through the use of the sample construction model, and then subsequently generates the required EEG signal samples based on the initial EEG signal samples and the expanded samples to complete the expansion of the EEG signal samples, which effectively enhances the diversity of the EEG signal samples, thereby facilitating the improvement of the generalization ability and accuracy of the anomaly detection model generated by subsequent training.

[0132] In some optional implementations of this embodiment, the test module includes:

[0133] The second acquisition submodule is used to input the test set into the specified anomaly detection model and obtain the performance indicator value of the specified anomaly detection model; wherein the performance indicator value is a value corresponding to a preset performance indicator, and the number of the performance indicators includes multiple.

[0134] In this embodiment, the test data set is input into the specified anomaly detection model for testing, and after the model test is completed, the performance indicator value of the specified anomaly detection model corresponding to the preset performance indicator is recorded. The above performance indicator specifically includes a recall rate indicator and an accuracy rate indicator. Correspondingly, the above performance indicator value includes a recall rate value and an accuracy rate value.

[0135] The second judgment submodule is used to judge whether all the performance indicator values ​​are within the corresponding preset value range.

[0136] In this embodiment, according to the actual model performance test requirements, a preset value range that meets normal expectations is pre-constructed for each performance indicator. Specifically, a recall value range that matches the recall indicator and an accuracy value range that matches the accuracy indicator are pre-constructed.

[0137] The first determination submodule is used to determine that the specified anomaly detection model has passed the test if all are within the corresponding preset value range.

[0138] In this embodiment, if all the performance indicator values ​​are within the corresponding preset value range, it indicates that the model effect (recall rate and accuracy) of the generated specified anomaly detection model has achieved the expected effect, and then the specified anomaly detection model will be judged to have passed the test.

[0139] The second determination submodule is used to determine that the specified anomaly detection model fails the test if the values ​​are not all within the corresponding preset value range.

[0140] In this embodiment, if at least one performance indicator value among all the performance indicator values ​​is not within the corresponding preset value range, it indicates that the model effect of the generated specified anomaly detection model does not achieve the expected effect, and then it is determined that the specified anomaly detection model has failed the test.

[0141] The present application inputs the test set into the specified anomaly detection model and obtains the performance indicator value of the specified anomaly detection model; then determines whether all the performance indicator values ​​are within the corresponding preset value range; if they are within the corresponding preset value range, the specified anomaly detection model is determined to have passed the test; and if they are not within the corresponding preset value range, the specified anomaly detection model is determined to have failed the test. The present application inputs the test set into the specified anomaly detection model and obtains the performance indicator value of the specified anomaly detection model, and then evaluates the performance indicator value of the obtained specified anomaly detection model based on the use of the preset value range, thereby accurately generating a performance test result corresponding to the specified anomaly detection model, thereby ensuring the accuracy of the obtained performance test result.

[0142] In some optional implementations of this embodiment, the above-mentioned cognitive impairment detection device based on artificial intelligence further includes:

[0143] The third generating module is used to generate a corresponding target detection report based on the mild cognitive impairment detection result.

[0144] In this embodiment, the specific implementation process of generating the corresponding target detection report based on the mild cognitive impairment detection result will be described in further detail in subsequent specific embodiments of the present application, and will not be elaborated in detail here.

[0145] The fourth acquisition module is used to acquire the communication information of the user.

[0146] In this embodiment, the communication information may refer to the user's mobile phone number or email address.

[0147] A push module is used to push the target detection report to the user based on the communication information.

[0148] In this embodiment, the target detection report may be sent to the communication terminal corresponding to the user according to the obtained communication information, so as to complete the push processing of the target detection report.

[0149] This application generates a corresponding target detection report based on the shallow cognitive impairment detection result; then obtains the communication information of the user; and subsequently pushes the target detection report to the user based on the communication information. After generating the shallow cognitive impairment detection result corresponding to the user, this application will intelligently generate a corresponding target detection report based on the shallow cognitive impairment detection result, thereby realizing the automatic generation of the target detection report, eliminating the need to manually perform the manual report generation operation, and improving the generation efficiency of the target detection report. In addition, the target detection report will also be pushed to the user based on the user's communication information, so that the user can promptly understand the detection status of his or her shallow cognitive impairment and perform corresponding subsequent processing, thereby improving the user's experience.

[0150] In some optional implementations of this embodiment, the third generating module includes:

[0151] The conversion submodule is used to convert the shallow cognitive impairment detection result into a target detection result corresponding to a preset understanding style.

[0152] In this embodiment, the above-mentioned understanding style refers to a non-professional style that is easy to understand. If the SCI detection result is that the user has SCI, the corresponding target detection result content may be: "According to our assessment, you are currently facing a certain risk of SCI." If the SCI detection result is that the user does not have SCI, the corresponding target detection result content may be: "Our assessment results show that you currently have no significant risk of SCI."

[0153] The third acquisition submodule is used to acquire medical advice information corresponding to the mild cognitive impairment detection result.

[0154] In this embodiment, if the result of the mild cognitive impairment detection is that the user has mild cognitive impairment, the corresponding medical advice information includes: "In view of your risk of mild cognitive impairment, we recommend that you undergo further medical evaluation to more accurately understand your cognitive condition." If the result of the mild cognitive impairment detection is that the user does not have mild cognitive impairment, the corresponding medical advice information includes: "We recommend that you perform regular cognitive health monitoring to promptly detect and address any potential problems."

[0155] The fourth acquisition submodule is used to obtain lifestyle recommendation information corresponding to the mild cognitive impairment detection result.

[0156] In this embodiment, if the result of the mild cognitive impairment detection is that the user has mild cognitive impairment, the corresponding lifestyle advice information includes: "At the same time, we recommend that you maintain a healthy lifestyle, which helps prevent and delay the development of cognitive impairment." If the result of the mild cognitive impairment detection is that the user does not have mild cognitive impairment, the corresponding lifestyle advice information includes: "Although you currently do not have a significant risk of mild cognitive impairment, we still recommend that you maintain a healthy lifestyle to prevent possible cognitive problems in the future."

[0157] The fifth acquisition submodule is used to obtain a preset test report template.

[0158] In this embodiment, the above-mentioned test report template may specifically be a template file constructed according to actual business needs, and the template file includes a test result area, a medical advice area, and a lifestyle advice area.

[0159] The filling submodule is used to fill the target detection result, the medical advice information and the lifestyle advice information into the detection report template to obtain the target detection report.

[0160] In this embodiment, the corresponding target detection report can be obtained by filling the target detection result, the medical advice information and the lifestyle advice information into the detection result area, the medical advice area and the lifestyle advice area in the detection report template respectively.

[0161] The present application converts the shallow cognitive impairment test result into a target test result corresponding to a preset understanding style; then obtains medical advice information corresponding to the shallow cognitive impairment test result; and obtains lifestyle advice information corresponding to the shallow cognitive impairment test result; then obtains a preset test report template; and subsequently fills the target test result, the medical advice information, and the test report template into the test report template to obtain the target test report. The present application converts the shallow cognitive impairment test result into a target test result corresponding to a preset understanding style, and obtains medical advice information corresponding to the shallow cognitive impairment test result; and obtains lifestyle advice information corresponding to the shallow cognitive impairment test result; and then fills the target test result, the medical advice information, and the lifestyle advice information into the preset test report template, thereby automatically and intelligently completing the construction of the target test report and improving the content richness of the target test report.

[0162] In some optional implementations of this embodiment, there are also some shallow cognitive impairment detection methods based on machine learning in the prior art. However, these detection methods lack automation and real-time processing capabilities, and most of them rely on fixed feature sets, which are difficult to adapt to individual differences and dynamically changing EEG signals. In addition, when processing large-scale, multi-dimensional EEG data, these methods often face the problems of dimensionality disaster and high computational complexity, resulting in unstable detection results and insufficient generalization capabilities. Finally, existing detection systems generally lack sensitivity to abnormal EEG activity and are unable to capture and identify weak early MCI signals in a timely manner.

[0163] This application combines the Transformer and Anomaly Attention methods to effectively solve the problems of existing methods: manual feature extraction is complex and time-consuming; it relies on a fixed feature set and is difficult to adapt to individual differences and dynamically changing EEG signals; and it faces the problems of dimensionality disaster and high computational complexity when processing large-scale, multi-dimensional EEG data. The Anomaly Transformer can automatically learn complex patterns and features in EEG signals through unified modeling of point representations and global associations of time series, reducing reliance on manual feature extraction. At the same time, the Anomaly Attention mechanism improves the sensitivity and accuracy of MCI detection by capturing subtle abnormal changes in EEG signals.

[0164] Specifically, the association weight distribution generated by the self-attention mechanism of Anomaly Transformer reflects the rich association between each time point and the entire sequence. Due to its rarity, it is difficult for anomalies to establish significant associations with the entire sequence, which makes the associations of anomalies mainly concentrated on their adjacent time points. This neighborhood concentration bias can effectively distinguish between normal and abnormal points. By introducing an anomaly detection criterion based on association differences, Anomaly Transformer uses the minimum and maximum strategy to amplify the distinction between normal and abnormal points, thereby realizing real-time and automatic processing of EEG signals and improving the generalization ability and robustness of the model. This method not only achieves advanced detection results on a variety of unsupervised time series anomaly detection benchmarks, but also can significantly improve the efficiency and accuracy of early detection of MCI.

[0165] This application can be applied in the field of smart medical care, thereby promoting the construction of smart cities.

[0166] It should be emphasized that in order to further ensure the privacy and security of the above-mentioned mild cognitive impairment test results, the above-mentioned mild cognitive impairment test results can also be stored in a blockchain node.

[0167] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0168] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0169] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0170] To solve the above technical problems, the present application also provides a computer device. Figure 3 , Figure 3 This is a basic structural block diagram of the computer device in this embodiment.

[0171] The computer device 3 includes a memory 31, a processor 32, and a network interface 33 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 3 with components 31-33, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0172] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0173] The memory 31 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 31 can be an internal storage unit of the computer device 3, such as a hard disk or memory of the computer device 3. In other embodiments, the memory 31 can also be an external storage device of the computer device 3, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. equipped on the computer device 3. Of course, the memory 31 can also include both the internal storage unit of the computer device 3 and its external storage device. In this embodiment, the memory 31 is generally used to store the operating system and various application software installed on the computer device 3, such as computer-readable instructions of a cognitive impairment detection device based on artificial intelligence. In addition, the memory 31 can also be used to temporarily store various types of data that have been output or are to be output.

[0174] The processor 32 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip in some embodiments. The processor 32 is generally used to control the overall operation of the computer device 3. In this embodiment, the processor 32 is used to run computer-readable instructions stored in the memory 31 or process data, such as computer-readable instructions for running the artificial intelligence-based cognitive impairment detection device.

[0175] The network interface 33 may include a wireless network interface or a wired network interface. The network interface 33 is generally used to establish a communication connection between the computer device 3 and other electronic devices.

[0176] In this embodiment, the functions of each module in the cognitive impairment detection device based on artificial intelligence as in the above-mentioned embodiment are realized when the processor executes the computer-readable instructions stored in the memory.

[0177] In this embodiment, the cognitive impairment detection device based on artificial intelligence first obtains the original EEG signal of the user; and preprocesses the original EEG signal to obtain the corresponding EEG signal sequence; then adds classification tags and position codes to the EEG signal sequence to obtain the corresponding target EEG signal sequence; then processes the target EEG signal sequence based on the abnormal detection layer in the trained abnormal detection model to obtain the target feature corresponding to the target EEG signal sequence; subsequently, the target feature is predicted and processed based on the fully connected layer in the abnormal detection model to obtain the corresponding prediction result; further obtains the preset threshold, and generates a numerical comparison result between the prediction result and the preset threshold; finally, based on the numerical comparison result, generates a shallow cognitive impairment detection result corresponding to the user. This application obtains the target EEG signal sequence by preprocessing the original EEG signal of the user obtained and adding classification tags and position codes, and then uses the trained abnormal detection model to predict the target EEG signal sequence to obtain the corresponding prediction result, and finally analyzes the prediction result based on the preset threshold, so as to quickly and accurately generate a shallow cognitive impairment detection result corresponding to the user. The present application adopts a method of detecting mild cognitive impairment based on an abnormality detection model, which no longer requires high reliance on manual processing and effectively improves the sensitivity and accuracy of mild cognitive impairment detection.

[0178] The present application also provides another embodiment, namely, providing a computer-readable storage medium, which stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the functions of each module in the above-mentioned artificial intelligence-based cognitive impairment detection device.

[0179] In this embodiment, the cognitive impairment detection device based on artificial intelligence first obtains the original EEG signal of the user; and preprocesses the original EEG signal to obtain the corresponding EEG signal sequence; then adds classification tags and position codes to the EEG signal sequence to obtain the corresponding target EEG signal sequence; then processes the target EEG signal sequence based on the abnormal detection layer in the trained abnormal detection model to obtain the target feature corresponding to the target EEG signal sequence; subsequently, the target feature is predicted and processed based on the fully connected layer in the abnormal detection model to obtain the corresponding prediction result; further obtains the preset threshold, and generates a numerical comparison result between the prediction result and the preset threshold; finally, based on the numerical comparison result, generates a shallow cognitive impairment detection result corresponding to the user. This application obtains the target EEG signal sequence by preprocessing the original EEG signal of the user obtained and adding classification tags and position codes, and then uses the trained abnormal detection model to predict the target EEG signal sequence to obtain the corresponding prediction result, and finally analyzes the prediction result based on the preset threshold, so as to quickly and accurately generate a shallow cognitive impairment detection result corresponding to the user. The present application adopts a method of detecting mild cognitive impairment based on an abnormality detection model, which no longer requires high reliance on manual processing and effectively improves the sensitivity and accuracy of mild cognitive impairment detection.

[0180] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0181] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A cognitive impairment detection device based on artificial intelligence, characterized in that: include: The first acquisition module is used to acquire the user's original EEG signal; A preprocessing module, used for preprocessing the original EEG signal to obtain a corresponding EEG signal sequence; A first processing module is used to add classification tags and position codes to the EEG signal sequence to obtain a corresponding target EEG signal sequence; A second processing module is used to process the target EEG signal sequence based on the anomaly detection layer in the trained anomaly detection model to obtain a target feature corresponding to the target EEG signal sequence; A prediction module, used to perform prediction processing on the target feature based on the fully connected layer in the anomaly detection model to obtain a corresponding prediction result; A first generating module, used to obtain a preset threshold value and generate a numerical comparison result between the prediction result and the preset threshold value; The second generating module is used to generate a mild cognitive impairment detection result corresponding to the user based on the numerical comparison result.

2. The artificial intelligence-based cognitive impairment detection device according to claim 1, characterized in that: The preprocessing module comprises: A filter submodule, used for filtering the original EEG signal to obtain a corresponding first EEG signal; A segmentation submodule, used to perform channel segmentation processing on the first EEG signal to obtain a corresponding second EEG signal; a flattening submodule, used for performing a data flattening operation on the second EEG signal to obtain a corresponding third EEG signal; A projection submodule, used for performing linear projection processing on the third EEG signal to obtain a corresponding fourth EEG signal; A determination submodule is used to use the fourth EEG signal as the target EEG signal.

3. The artificial intelligence-based cognitive impairment detection device according to claim 1, characterized in that: The second generation module comprises: A first judgment submodule, used to judge whether the numerical comparison result is that the prediction result is greater than the preset threshold; A first generating submodule, configured to generate a first mild cognitive impairment detection result indicating that the user has mild cognitive impairment if the predicted result is greater than the preset threshold; The second generating submodule is configured to generate a second mild cognitive impairment detection result indicating that the user does not have mild cognitive impairment if the predicted result is less than the preset threshold.

4. The artificial intelligence-based cognitive impairment detection device according to claim 1, characterized in that: The artificial intelligence-based cognitive impairment detection device also includes: The second acquisition module is used to acquire pre-constructed EEG signal samples; A division module, used for dividing the EEG signal samples into a training set and a test set; A third acquisition module is used to acquire a preset initial anomaly detection model and initialize the initial anomaly detection model; A third processing module, configured to input the training set into the initial anomaly detection model for forward propagation, and calculate an error between a model output of the initial anomaly detection model and a true label of the training set; A fourth processing module, configured to calculate a gradient based on the error, update a model weight of the initial anomaly detection model through a back propagation algorithm, and optimize the error using a minimization strategy and a maximization strategy; A fifth processing module is used to repeat the forward propagation and back propagation processes until a preset iteration condition is met to obtain a corresponding specified anomaly detection model; A testing module, used for testing the specified anomaly detection model based on the test set; A determination module is used to use the specified anomaly detection model as the anomaly detection model if the specified anomaly detection model passes the test.

5. The artificial intelligence-based cognitive impairment detection device according to claim 4, characterized in that: The second acquisition module includes: The first acquisition submodule is used to acquire the pre-collected initial EEG signal samples; The calling submodule is used to call the preset sample to build the model; A construction submodule, used for performing sample construction processing on the initial EEG signal sample based on the sample construction model to obtain a corresponding expanded sample; The third generating submodule is used to generate the EEG signal sample based on the initial EEG signal sample and the expanded sample.

6. The artificial intelligence-based cognitive impairment detection device according to claim 4, characterized in that: The test module includes: A second acquisition submodule is used to input the test set into the specified anomaly detection model and obtain a performance indicator value of the specified anomaly detection model; wherein the performance indicator value is a value corresponding to a preset performance indicator, and the number of the performance indicators includes a plurality; A second judgment submodule is used to judge whether all the performance indicator values ​​are within the corresponding preset value range; A first determination submodule, configured to determine that the specified anomaly detection model passes the test if all values ​​are within a corresponding preset value range; The second determination submodule is used to determine that the specified anomaly detection model fails the test if the values ​​are not all within the corresponding preset value range.

7. The artificial intelligence-based cognitive impairment detection device according to claim 1, characterized in that: The artificial intelligence-based cognitive impairment detection device also includes: A third generating module, configured to generate a corresponding target detection report based on the mild cognitive impairment detection result; A fourth acquisition module, used to acquire the communication information of the user; A push module is used to push the target detection report to the user based on the communication information.

8. The artificial intelligence-based cognitive impairment detection device according to claim 7, characterized in that: The third generation module comprises: A conversion submodule, used for converting the shallow cognitive impairment detection result into a target detection result corresponding to a preset understanding style; A third acquisition submodule is used to acquire medical advice information corresponding to the mild cognitive impairment test result; A fourth acquisition submodule is used to acquire lifestyle recommendation information corresponding to the mild cognitive impairment test result; The fifth acquisition submodule is used to obtain a preset test report template; The filling submodule is used to fill the target detection result, the medical advice information and the lifestyle advice information into the detection report template to obtain the target detection report.

9. A computer device, characterized in that: It includes a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the functions of each module in the artificial intelligence-based cognitive impairment detection device as described in any one of claims 1 to 8 are realized.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the functions of each module in the artificial intelligence-based cognitive impairment detection device as described in any one of claims 1 to 8.

Images