Cognitive disorder typing diagnosis method based on multi-modal data of electroencephalogram and heart rate
By combining multimodal data acquisition and processing of EEG and heart rate, the problem of insufficient accuracy in cognitive impairment classification in existing technologies has been solved, enabling accurate auxiliary diagnosis of cognitive impairment and supporting doctors in the specific classification of cognitive function impairment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- JIANGSU BOYA TECH CO LTD
- Filing Date
- 2026-04-02
- Publication Date
- 2026-05-12
AI Technical Summary
Existing technologies cannot continuously and comprehensively collect multimodal data in natural life scenarios, lack the ability to continuously perceive the real state, resulting in insufficient accuracy of cognitive impairment classification and inability to achieve synchronous and effective real-time diagnosis.
A multimodal data acquisition method based on EEG and heart rate was adopted. EEG caps and ECG electrodes were used to acquire patients' EEG and ECG data. Combined with environmental stimulus data, a multimodal dataset was generated. Significant potential changes were screened by setting thresholds, and the percentage of potential datasets was calculated to assist in the diagnosis of cognitive impairment classification.
It provides accurate auxiliary diagnosis for the classification of cognitive impairment, enabling better assessment of cognitive function impairment, supporting doctors in classifying mild, moderate and severe cognitive impairment, and improving the synchronicity and effectiveness of diagnosis.
Smart Images
Figure CN122004786A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a cognitive impairment classification and diagnosis method based on multimodal data of electroencephalography and heart rate, belonging to the field of cognitive impairment evaluation technology. Background Technology
[0002] With the accelerating aging of the global population, cognitive impairment has become a major public health challenge affecting the health and quality of life of the elderly. Cognitive impairment not only places a heavy burden of care and economic pressure on patients and their families, but also poses a severe test to social medical resources. Early identification and effective intervention for cognitive impairment are key to slowing disease progression and improving patient prognosis. The incidence of cognitive impairment continues to rise with the increasing global aging population. Currently, there is no cure, and early intervention to slow disease progression has become a core clinical strategy. Numerous non-pharmacological interventions (such as cognitive training, exercise therapy, and neuromodulation) and pharmacological intervention programs have emerged, but how to scientifically quantify the effectiveness of interventions remains a challenge for the industry.
[0003] With the development of computer technology and neuroscience, brain-computer interface (BCI) technology is being increasingly widely applied in assistive rehabilitation therapy for cognitive disorders such as autism spectrum disorder, improving attention levels. A BCI is a technology that converts neural signals from the brain into operable command signals to achieve human-computer interaction. In monitoring and assessing a patient's attention, BCI technology can objectively monitor and quantify attention, thus more accurately assessing changes in attention. For example, BCIs can record a patient's electroencephalogram (EEG) signals, and these signals can be analyzed to determine the patient's attentional state.
[0004] Existing technologies in the field of cognitive impairment classification, especially in terms of specific classification of impaired cognitive function, cannot continuously and comprehensively collect multimodal data from users in natural life scenarios. They lack the ability to continuously perceive the real state from the perspective of data analysis. The assessment and task selection do not achieve patient-environment adaptation, resulting in insufficient accuracy of cognitive impairment classification. After general classification, there is a lack of quantitative effect verification and closed-loop optimization, which cannot help doctors to achieve real-time judgment on the synchronicity and effectiveness of cognitive impairment classification diagnosis. Summary of the Invention
[0005] The purpose of this invention is to overcome the technical defects of the existing technology, solve the above-mentioned technical problems, and propose a cognitive impairment classification and diagnosis method based on multimodal data of EEG and heart rate, so as to realize the auxiliary classification and diagnosis of cognitive impairment from the perspective of cognitive data analysis.
[0006] The present invention specifically adopts the following technical solution: a cognitive impairment classification and diagnosis method based on multimodal data of electroencephalography and heart rate, comprising the following steps: Step SS1: Place the EEG cap on the patient's brain and attach the ECG electrodes to the patient's abdominal skin. The output ends of the EEG cap and the ECG electrodes are connected to the EEG acquisition computer through a signal adapter board and a signal amplifier. The stimulation presentation computer is connected to the signal amplifier through a synchronization box. Step SS2: The stimulation presentation computer synchronously triggers or deactivates the EEG cap and the ECG electrodes; the EEG cap collects the patient's EEG data and outputs it to the EEG acquisition computer; the ECG electrodes collect the patient's ECG data and output it to the EEG acquisition computer. Step SS3: Based on the obtained EEG and ECG data, environmental stimulus data are simultaneously superimposed to generate a multimodal dataset. The multimodal dataset is processed, and the percentage of potential datasets exceeding a set threshold is output as a percentage of the total number of environmental stimulus-related potential data.
[0007] In a preferred embodiment, the environmental stimulus data includes relevant event data, irrelevant event data, and quasi-event data, and the event-related potential (ERP) data, which is the change in relevant potentials caused by presenting or withdrawing a certain stimulus event, output by the EEG acquisition computer to the patient.
[0008] In a preferred embodiment, the EEG cap collects the secondary current generated by the primary current produced by the patient's neurons, and outputs a waveform diagram of the voltage generated on the scalp surface over time through an EEG acquisition computer.
[0009] In a preferred embodiment, the ECG electrode outputs the potential transmission of the patient's heart, and the computer outputs waveforms of the time nodes and intensity of the heartbeat signal sequence through electroencephalography (EEG) acquisition.
[0010] In a preferred embodiment, step SS3 includes: Based on relevant event data, irrelevant event data, and benchmark event data from environmental stimulus data, the EEG and ECG data of a patient are output simultaneously by overlaying the relevant event data; the EEG and ECG data of the patient are output simultaneously by overlaying the irrelevant event data; and the EEG and ECG data of the patient are output simultaneously by overlaying the benchmark event data. Then, these data are compared with the original EEG and ECG data obtained at the same time without environmental stimulus data, and the relevant event potential data △ERP1, irrelevant event potential data △ERP2, and benchmark event potential data △ERP3 are output to show the potential changes over time.
[0011] In a preferred embodiment, step SS3 further includes: EEG data 1 for relevant event data is denoted as set VG1(1, ..., x), where x is the number of relevant event questions, and VG1(x) is the EEG voltage signal of the x-th relevant event question; ECG data 1 for relevant event data is denoted as set HG1(1, ..., x), where x is the number of relevant event questions, and HG1(x) is the ECG voltage signal of the x-th relevant event question; EEG data 2 for irrelevant event data is denoted as set VG2(1, ..., y), where y is the number of irrelevant event questions, and VG2(y) is the EEG voltage signal of the y-th irrelevant event question; ECG data 2 for irrelevant event data is denoted as set HG2(1, ..., y), where... Let y be the number of irrelevant event questions, and HG2(y) be the ECG voltage signal of the y-th irrelevant event question; let the EEG data EEG3 for the standard event data be denoted as set VG3(1,…,z), where z is the number of standard event questions, and VG3(z) be the EEG voltage signal of the z-th standard event question; let the ECG data ECG3 for the standard event data be denoted as set HG3(1,…,z), where z is the number of standard event questions, and HG3(z) be the ECG voltage signal of the z-th standard event question; let the raw EEG data EEG and ECG data ECG at the same time without applied environmental stimuli be denoted as EEG0 and ECG0, respectively. Then the relevant event potential data ΔERP1, the irrelevant event potential data ΔERP2, and the standard event potential data ΔERP3 are respectively: ; ; ; Relevant event potential data ERP1, Irrelevant Event Potential Data ERP2, Quasi-event potential data In ERP3, the portions exceeding the set thresholds PMax1, PMax2, and PMax3 constitute a new set. ERP1X{1,…,i}, ERP2Y{1, ...,j}, ERP3Z{1,…,k}, where i ≤ x, j ≤ y, k ≤ z, PMax1 is the permissible potential change threshold for relevant event stimuli, PMax2 is the permissible potential change threshold for irrelevant event stimuli, and PMax3 is the permissible potential change threshold for quasi-event stimuli.
[0012] In a preferred embodiment, step SS3 further includes: obtaining the percentage of the number of potential datasets exceeding a set threshold relative to the total number of potential datasets of environmental stimuli, including: ; ; ; ; Wherein, C1 is the percentage of the number of relevant event potential datasets exceeding the set threshold out of the total potential dataset of all relevant event stimuli; C2 is the percentage of the number of irrelevant event potential datasets exceeding the set threshold out of the total potential dataset of all irrelevant event stimuli; C3 is the percentage of the number of benchmark event potential datasets exceeding the set threshold out of the total potential dataset of all benchmark event stimuli; C is the percentage of the number of potential datasets exceeding the set threshold out of the total potential dataset of all environmental data stimuli; the larger the percentage, the more obvious the patient's cognition of environmental data stimuli, and relatively speaking, the milder the degree of cognitive impairment.
[0013] In a preferred embodiment, the signal adapter board has 8-128 channels.
[0014] In a preferred embodiment, the signal amplifier is a 128-channel amplifier.
[0015] As a preferred embodiment, it also includes an eye tracker, which is connected to a stimulation presentation computer via a synchronization box, and the eye tracker collects the patient's eye movement trajectory data and outputs it to the EEG acquisition computer.
[0016] The beneficial effects achieved by this invention are as follows: This invention addresses the shortcomings of existing technologies in the field of cognitive impairment classification, particularly in the specific classification of impaired cognitive function. These shortcomings include the inability to continuously and comprehensively collect multimodal data from users in natural life scenarios, a lack of continuous perception of the real-world state from a data analysis perspective, failure to adapt assessment and task selection to the patient and environment, resulting in insufficient accuracy in cognitive impairment classification, and a lack of quantitative effect verification and closed-loop optimization after general classification. Furthermore, these technologies fail to assist physicians in achieving real-time assessment of the synchronicity and effectiveness of cognitive impairment classification diagnosis. This invention proposes a method based on… A method for classifying and diagnosing cognitive impairment using multimodal data of electroencephalography (EEG) and heart rate includes the following steps: Step SS1: An EEG cap is placed on the patient's brain, and ECG electrodes are attached to the patient's abdominal skin. The output terminals of the EEG cap and ECG electrodes are connected to an EEG acquisition computer via a signal adapter and a signal amplifier. A stimulation presentation computer is connected to the signal amplifier via a synchronization box. Step SS2: The stimulation presentation computer synchronously triggers or deactivates the connection between the EEG cap and the ECG electrodes. The EEG cap acquires the patient's EEG data and outputs it to the EEG computer. The computer collects ECG data from the patient; the ECG electrodes collect the patient's ECG data and output it to the EEG computer; Step SS3: Based on the obtained EEG data and ECG data, environmental stimulus data is simultaneously superimposed to generate a multimodal dataset. The multimodal dataset is processed, and the percentage of potential datasets exceeding a set threshold is output as the percentage of the total number of potential datasets from environmental stimuli. The technical solution and processed data of this invention can serve as auxiliary data support for doctors to diagnose patients' cognitive impairment classification, such as mild cognitive impairment, moderate cognitive impairment, and severe cognitive impairment. Furthermore, it provides auxiliary data support for doctors to specifically classify cognitive function impairment. Overall, it solves the technical defects of existing technologies in the field of cognitive impairment classification, especially in the specific classification of impaired cognitive function. These defects include the inability to continuously and comprehensively collect multimodal data from users in natural life scenarios, the lack of continuous perception of the real state from the perspective of data analysis, the failure to achieve patient-environment adaptation in assessment and task selection, resulting in insufficient accuracy of cognitive impairment classification, and the lack of quantitative effect verification and closed-loop optimization after general classification, which cannot assist doctors in achieving real-time judgment of the synchronicity and effectiveness of cognitive impairment classification diagnosis. Attached Figure Description
[0017] Figure 1 This is a flowchart of the cognitive impairment classification and diagnosis method based on multimodal data of EEG and heart rate of the present invention.
[0018] Figure 2 This is a schematic diagram of a preferred embodiment of the present invention. Detailed Implementation
[0019] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.
[0020] Example 1: As Figure 1 and Figure 2 As shown, this invention proposes a cognitive impairment classification and diagnostic method based on multimodal data of EEG and heart rate, including the following steps: Step SS1: Place the EEG cap on the patient's brain and attach the ECG electrodes to the patient's abdominal skin. The output ends of the EEG cap and the ECG electrodes are connected to the EEG acquisition computer through a signal adapter board and a signal amplifier. The stimulation presentation computer is connected to the signal amplifier through a synchronization box. Step SS2: The stimulation presentation computer synchronously triggers or deactivates the EEG cap and the ECG electrodes; the EEG cap collects the patient's EEG data and outputs it to the EEG acquisition computer; the ECG electrodes collect the patient's ECG data and output it to the EEG acquisition computer; wherein, the ECG data collected by the ECG electrodes utilizes electrodes attached to the surface of the skin to detect the electrical potential transmission of the patient's heart, and the EEG cap collects data from the patient's brain, specifically from the prefrontal, posterior, parietal, occipital, and temporal lobes of the cerebral cortex. The stimulation signals of the lobes are as follows: the prefrontal lobe is associated with mental functions such as communication and management, planning and judgment, creative leadership, and goal setting; the posterior frontal lobe is associated with cognitive functions such as logical reasoning, language function, spatial imagination, and conceptualization; the parietal lobe is associated with somatosensory functions such as somatosensory recognition, operational comprehension, somatosensory experience, and appreciation of crafts; the occipital lobe is associated with visual functions such as visual recognition, observation and comprehension, visual experience, and image appreciation; and the temporal lobe is associated with auditory functions such as auditory recognition, language comprehension, auditory experience, and music appreciation. Step SS3: Based on the obtained EEG and ECG data, environmental stimulus data are simultaneously superimposed to generate a multimodal dataset. The multimodal dataset is processed, and the percentage of potential datasets exceeding a set threshold is output as a percentage of the total number of potential datasets from environmental stimuli.
[0021] Optionally, the environmental stimulus data includes relevant event data, irrelevant event data, and benchmark event data. Event-related potentials (ERPs) are the changes in potentials triggered by presenting or withdrawing a certain stimulus event, output by the EEG acquisition computer to the patient. Relevant event data directly addresses the core of the event, such as "Did you steal money from the office last night?" or "Did you have dinner with someone last night?". Irrelevant event data refers to everyday event stimulus data unrelated to the event, such as "Did you forget to eat breakfast today?", used to establish a baseline for physiological responses. Benchmark event data refers to stimulus questions designed to make even honest people nervous, such as "Have you done anything illegal?", used to compare the intensity of responses to relevant stimuli.
[0022] Optionally, the EEG cap collects the secondary current generated by the primary current produced by the patient's neurons, and outputs a waveform diagram of the voltage generated on the scalp surface over time through an EEG acquisition computer.
[0023] Optionally, the ECG electrode outputs the potential transmission of the patient's heart, and the computer outputs waveforms of the time nodes and intensity of the heartbeat signal sequence through electroencephalogram acquisition.
[0024] Optionally, step SS3 includes: Based on relevant event data, irrelevant event data, and benchmark event data from environmental stimulus data, the system simultaneously overlays relevant event data to output a patient's EEG1 and ECG1 data; simultaneously overlays irrelevant event data to output the patient's EEG2 and ECG2 data; and simultaneously overlays benchmark event data to output the patient's EEG3 and ECG3 data. These data are then compared with the original EEG and ECG data obtained at the same time or period without applied environmental stimulus data to output relevant event potential data showing potential changes over time. ERP1, Irrelevant Event Potential Data ERP2, Quasi-event potential data ERP3.
[0025] Specifically, the EEG data (EEG1) for relevant event data is denoted as set VG1(1,…,x), where x is the number of relevant event questions, and VG1(x) is the EEG voltage signal of the x-th relevant event question. The ECG data (ECG1) for relevant event data is denoted as set HG1(1,…,x), where x is the number of relevant event questions, and HG1(x) is the ECG voltage signal of the x-th relevant event question. The EEG data (EEG2) for irrelevant event data is denoted as set VG2(1,…,y), where y is the number of irrelevant event questions, and VG2(y) is the EEG voltage signal of the y-th irrelevant event question. The ECG data (ECG2) for irrelevant event data is denoted as set HG2(…,y). HG2(y) is the electrocardiogram (ECG) voltage signal of the y-th irrelevant event problem; the EEG data EEG3 for the standard event data is denoted as set VG3(1,…,z), where z is the number of standard event problems, and VG3(z) is the EEG voltage signal of the z-th standard event problem; the ECG data ECG3 for the standard event data is denoted as set HG3(1,…,z), where z is the number of standard event problems, and HG3(z) is the ECG voltage signal of the z-th standard event problem; the raw EEG data EEG and ECG data ECG at the same time without applied environmental stimuli are denoted as EEG0 and ECG0, respectively, then the relevant event potential data... ERP1, Irrelevant Event Potential Data ERP2, Quasi-event potential data ERP3 are as follows: ; ; ; Relevant event potential data ERP1, Irrelevant Event Potential Data ERP2, Quasi-event potential data In ERP3, the portions exceeding the set thresholds PMax1, PMax2, and PMax3 constitute a new set. ERP1X{1,…,i}, ERP2Y{1, ...,j}, ERP3Z{1, …, k}. Where i ≤ x, j ≤ y, k ≤ z, PMax1 is the permissible potential change threshold for relevant event stimuli, PMax2 is the permissible potential change threshold for irrelevant event stimuli, and PMax3 is the permissible potential change threshold for quasi-event stimuli. The values of PMax1, PMax2, and PMax3 are determined based on factors such as the patient's age and weight, and are not the same for all patients.
[0026] Obtain the percentage of potential datasets exceeding the set threshold out of the total number of environmental stimulus potential datasets, including: ; ; ; ; Wherein, C1 is the percentage of relevant event potential datasets exceeding the set threshold out of the total potential dataset of all relevant event stimuli; C2 is the percentage of irrelevant event potential datasets exceeding the set threshold out of the total potential dataset of all irrelevant event stimuli; C3 is the percentage of benchmark event potential datasets exceeding the set threshold out of the total potential dataset of all benchmark event stimuli; and C is the percentage of potential datasets exceeding the set threshold out of the total potential dataset of all environmental data stimuli. The higher the percentage, the more pronounced the patient's cognition of environmental data stimuli, and relatively milder the degree of cognitive impairment. This data serves as supplementary data support for doctors to diagnose the classification of patients' cognitive impairment, such as mild, moderate, and severe cognitive impairment. Furthermore, the above data provides supplementary data support for doctors to classify specific areas of cognitive function impairment. Based on the impaired cognitive function area, the classification includes: memory impairment (A), aphasia (B), agnosia (C), apraxia (D), visuospatial impairment (E), executive function impairment (F), and computational impairment (G). Among them, memory impairment A is a decline, forgetting, or error in recent or long-term memory; aphasia B is difficulty in language expression or comprehension; agnosia C is loss of the ability to recognize common objects, people, or sounds; apraxia D is inability to perform learned complex actions; visuospatial impairment E is difficulty in spatial orientation and visual discrimination; executive function impairment F is a decline in higher thinking abilities such as planning and organization; and calculation impairment G is impaired ability to perform simple calculations.
[0027] Optionally, the signal adapter board has 8-128 channels.
[0028] Optionally, the signal amplifier is a 128-channel amplifier.
[0029] It should be noted that the EEG data of this invention needs to be preprocessed, which may specifically include: 1) Bandpass filtering: Digital filters retain signals between 0.5Hz and 40Hz; Notch filtering: Removes power frequency interference at 50Hz or 60Hz; 2) Remove bad segments: An automated algorithm is used to scan the entire continuous EEG data. If the amplitude of a certain segment of data exceeds the threshold, it is considered that the data is beyond repair, marked as a "bad segment", and removed from subsequent analysis. 3) Artifact removal: Using independent component analysis, the multi-channel EEG signal is decomposed into N statistically independent source signal components. The topographic maps and temporal waveforms of these independent components are examined. After identifying the components representing eye movements, blinking, or heartbeats, the weights of these components are set to zero. Finally, the remaining components representing pure brain activity are remixed to restore a clean multi-channel EEG signal. 4) Segmented extraction: The data is segmented into Epochs, with the stimulus presentation time in the task as time point 0; 5) Baseline correction: For each epoch, calculate the average voltage value during the pre-stimulus presentation period, and then subtract this average value from each data point of that epoch; 6) Frequency band energy ratio calculation: Apply Fast Fourier Transform or Wavelet Transform to each clean epoch to transform it from the time domain to the frequency domain, obtain the power spectral density, and then calculate it according to the preset frequency band range. Calculate the absolute power within each frequency band, calculate the total power = the sum of the absolute power of all frequency bands, and calculate the final data = (absolute power of a certain frequency band / total power). This calculation is performed on 100% of each frequency band, resulting in a set of energy percentage values, and thus the values of δ, θ, α, and β in... The proportion of energy in the frequency band.
[0030] Optionally, an eye tracker may also be included. This eye tracker is connected to a stimulus presentation computer via a synchronization box. The eye tracker collects the patient's eye movement trajectory data and outputs it to the EEG acquisition computer. Modern eye trackers generally consist of four systems: an optical system, a pupil center coordinate extraction system, a visual scene and pupil coordinate overlay system, and an image and data recording and analysis system. There are three basic types of eye movements: fixation, saccades, and pursuit movements. Eye movements can reflect the selection patterns of visual information and are of great significance for revealing the psychological mechanisms of cognitive processing. According to research reports, the data or parameters commonly used in psychological research using eye trackers mainly include: fixation trajectory diagrams, eye movement time, average velocity and distance (or amplitude) of saccade direction, pupil size (area or diameter, unit: pixel), and blink.
[0031] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0032] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 The computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0033] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.
Claims
1. A cognitive impairment classification and diagnostic method based on multimodal data of EEG and heart rate, characterized in that, Includes the following steps: Step SS1: Place the EEG cap on the patient's brain and attach the ECG electrodes to the patient's abdominal skin. The output ends of the EEG cap and the ECG electrodes are connected to the EEG acquisition computer through a signal adapter board and a signal amplifier. The stimulation presentation computer is connected to the signal amplifier through a synchronization box. Step SS2: The stimulation presentation computer synchronously triggers or deactivates the EEG cap and the ECG electrodes; the EEG cap collects the patient's EEG data and outputs it to the EEG acquisition computer; the ECG electrodes collect the patient's ECG data and output it to the EEG acquisition computer. Step SS3: Based on the obtained EEG and ECG data, environmental stimulus data are simultaneously superimposed to generate a multimodal dataset. The multimodal dataset is processed, and the percentage of potential datasets exceeding a set threshold is output as a percentage of the total number of environmental stimulus-related potential data.
2. The cognitive impairment classification and diagnosis method based on multimodal data of EEG and heart rate according to claim 1, characterized in that, The environmental stimulus data includes relevant event data, irrelevant event data, and standard event data. The event-related potential (ERP) data is generated when a certain stimulus event is presented or withdrawn by the EEG acquisition computer and output to the patient.
3. The cognitive impairment classification and diagnosis method based on multimodal data of EEG and heart rate according to claim 1, characterized in that, The EEG cap collects the secondary current generated by the primary current produced by the patient's neurons, and the EEG acquisition computer outputs a waveform diagram of the voltage generated on the scalp surface over time.
4. The cognitive impairment classification and diagnosis method based on multimodal data of EEG and heart rate according to claim 1, characterized in that, The ECG electrode outputs the patient's heart potential transmission, and the computer outputs waveforms of the time nodes and intensity of the heartbeat signal sequence through electroencephalography (EEG) acquisition.
5. The cognitive impairment classification and diagnosis method based on multimodal data of EEG and heart rate according to claim 1, characterized in that, Step SS3 includes: Based on relevant event data, irrelevant event data, and benchmark event data from environmental stimulus data, the system simultaneously overlays relevant event data to output a patient's EEG1 and ECG1 data; simultaneously overlays irrelevant event data to output the patient's EEG2 and ECG2 data; and simultaneously overlays benchmark event data to output the patient's EEG3 and ECG3 data. These data are then compared with the original EEG and ECG data obtained at the same time without applied environmental stimulus data to output relevant event potential data showing potential changes over time. ERP1, Irrelevant Event Potential Data ERP2, Quasi-event potential data ERP3.
6. The cognitive impairment classification and diagnosis method based on multimodal data of EEG and heart rate according to claim 5, characterized in that, Step SS3 specifically further includes: EEG data 1 for relevant event data is denoted as set VG1(1,…,x), where x is the number of relevant event questions, and VG1(x) is the EEG voltage signal of the x-th relevant event question; ECG data 1 for relevant event data is denoted as set HG1(1,…,x), where x is the number of relevant event questions, and HG1(x) is the ECG voltage signal of the x-th relevant event question; EEG data 2 for irrelevant event data is denoted as set VG2(1,…,y), where y is the number of irrelevant event questions, and VG2(y) is the EEG voltage signal of the y-th irrelevant event question; ECG data 2 for irrelevant event data is denoted as set HG2(1,…,y), where y is the number of irrelevant event questions. The number of events is given by HG2(y), where HG2(y) is the ECG voltage signal of the y-th irrelevant event event. The EEG data (EEG3) for the standard event data is denoted as set VG3(1,…,z), where z is the number of standard event events, and VG3(z) is the EEG voltage signal of the z-th standard event event. The ECG data (ECG3) for the standard event data is denoted as set HG3(1,…,z), where z is the number of standard event events, and HG3(z) is the ECG voltage signal of the z-th standard event event. The raw EEG and ECG data at the same time without environmental stimulation are denoted as EEG0 and ECG0, respectively. Then, the relevant event potential data ΔERP1, irrelevant event potential data ΔERP2, and standard event potential data ΔERP3 are respectively: ； ； ； Relevant event potential data ERP1, Irrelevant Event Potential Data ERP2, Quasi-event potential data In ERP3, the portions exceeding the set thresholds PMax1, PMax2, and PMax3 constitute a new set. ERP1X{1,…,i}, △ERP2Y{1,…,j}, △ERP3Z{1,…,k}, where i ≤ x, j ≤ y, k ≤ z, PMax1 is the permissible potential change threshold for relevant event stimuli, PMax2 is the permissible potential change threshold for irrelevant event stimuli, and PMax3 is the permissible potential change threshold for quasi-event stimuli.
7. The cognitive impairment classification and diagnosis method based on multimodal data of EEG and heart rate according to claim 6, characterized in that, Step SS3 further includes: obtaining the percentage of the number of correlated potential datasets exceeding a set threshold relative to the total number of environmental stimulus-related potential datasets, including: ； ； ； ； Wherein, C1 is the percentage of relevant event potential datasets exceeding the set threshold out of the total relevant event potential dataset; C2 is the percentage of irrelevant event potential datasets exceeding the set threshold out of the total irrelevant event potential dataset; C3 is the percentage of standard event potential datasets exceeding the set threshold out of the total standard event potential dataset; and C is the percentage of relevant potential datasets exceeding the set threshold out of the total environmental data stimulus dataset. The higher the percentage, the more obvious the patient's cognition of environmental data stimuli, and relatively speaking, the milder the degree of cognitive impairment.
8. The cognitive impairment classification and diagnosis method based on multimodal data of EEG and heart rate according to claim 1, characterized in that, The signal adapter board has 8-128 channels.
9. The cognitive impairment classification and diagnostic method based on multimodal data of EEG and heart rate according to claim 1, characterized in that, The signal amplifier is a 128-channel amplifier.
10. The cognitive impairment classification and diagnostic method based on multimodal data of EEG and heart rate according to claim 1, characterized in that, It also includes an eye tracker, which is connected to a stimulation presentation computer via a synchronization box. The eye tracker collects the patient's eye movement trajectory data and outputs it to the EEG acquisition computer.