Elderly psychological state assessment and early warning method and system

Abstract

The invention discloses an elderly psychological state assessment early warning method and system, and relates to the technical field of elderly psychological health monitoring, and the method comprises the steps: collecting electroencephalogram signals of an elderly assessment object in a daily fixed time period, calculating the power spectral density ratio of the electroencephalogram signals, and constructing a personal reference feature set of the elderly assessment object; obtaining a current electroencephalogram signal of the aged evaluation object, calculating a power spectral density ratio of the current electroencephalogram signal, comparing the power spectral density ratio with the personal reference feature set, and determining a short-term fluctuation amplitude and an accumulated deviation amount; and judging the abnormal degree of the psychological state according to the short-term fluctuation amplitude and the accumulated deviation, and generating a graded early warning signal containing a risk grade and a timestamp. According to the invention, the detection accuracy and early warning timeliness of the abnormal psychological state of the elderly are improved, and a further solution is provided for psychological health management of the elderly.

Description

Technical Field

[0001] The present invention relates to the technical field of mental health monitoring for the elderly, and specifically to a method and system for evaluating and warning the mental state of the elderly. Background Art

[0002] Mental health problems of the elderly have become an important challenge in the process of global population aging. Neuroscientific research shows that changes in cognitive function and emotional state of the elderly often precede changes in electroencephalogram (EEG) physiological characteristics in behavioral symptoms. Traditional EEG assessment methods mainly rely on clinical-level EEG devices in a hospital examination environment, and evaluate the absolute power or relative power of δ 、 θ 、 α and β bands. These methods have been widely used in brain cognitive function research and neuropsychiatric disease diagnosis. For example, a decrease in α wave activity is used to judge a decline in cognitive function, the change in the θ / α ratio is used to evaluate the attention level, and the change in the δ / β ratio is used to detect emotional abnormalities. However, these assessment methods are mainly designed for the diagnosis of specific diseases, lack sensitivity to changes in the subclinical state, and are difficult to meet the needs of daily monitoring and early warning.

[0003] There are three main limitations in the existing mental state assessment technologies for the elderly: First, traditional assessments mostly establish reference values based on group standards, ignoring significant individual differences within the elderly population, resulting in a lack of personalization in assessment results; second, most existing methods adopt single cross-sectional assessments and lack a dynamic monitoring mechanism, unable to capture progressive change trends; third, the signal preprocessing technology is not fine enough, and the ability to process noise and artifacts is limited. Especially when the quality of EEG data collected in a daily environment is low, the accuracy drops significantly. These problems make it difficult for existing technologies to achieve early and accurate warning of changes in the mental state of the elderly, missing the best opportunity for early intervention in mental problems. In addition, existing technologies mostly focus on the diagnosis of specific diseases rather than a full-spectrum assessment from the perspective of health management, and are difficult to meet the needs of preventive medicine. Summary of the Invention

[0004] In view of the above problems, the present invention is proposed.

[0005] Therefore, the present invention provides a method and system for evaluating and warning the mental state of the elderly, which can solve the problems mentioned in the background art.

[0006] To solve the above technical problems, the present invention provides the following technical solution: An elderly psychological state assessment and early warning method, comprising: collecting electroencephalogram (EEG) signals of an elderly assessment object at a fixed time period every day, calculating the power spectral density ratio of the EEG signals, and constructing a personal reference feature set of the elderly assessment object; obtaining the current EEG signals of the elderly assessment object, calculating the power spectral density ratio of the current EEG signals, comparing the power spectral density ratio with the personal reference feature set, and determining the short-term fluctuation amplitude and the cumulative deviation; determining the degree of abnormal psychological state according to the short-term fluctuation amplitude and the cumulative deviation, and generating a graded early warning signal including a risk level and a timestamp.

[0007] As a preferred embodiment of the elderly psychological state assessment and early warning method of the present invention, wherein: the calculation of the power spectral density ratio of the EEG signals includes: Preprocessing the EEG signals by a weighted symmetric singular spectrum analysis method; Calculating the power spectral density ratios of the δ band, θ band, α band, and β band of the preprocessed EEG signals.

[0008] As a preferred embodiment of the elderly psychological state assessment and early warning method of the present invention, wherein: the weighted symmetric singular spectrum analysis method includes: Embedding the EEG signals to form a trajectory matrix and performing singular value decomposition on the trajectory matrix to obtain eigenvalues and eigenvectors; Performing weighted and symmetric processing on the eigenvalues and the eigenvectors according to the inter-band phase synchronization index.

[0009] As a preferred embodiment of the elderly psychological state assessment and early warning method of the present invention, wherein: the calculation of the inter-band phase synchronization index includes: Using Hilbert transform to extract the instantaneous phase of each band of the EEG signals; Calculating the phase difference distribution between the instantaneous phases of each band and statistically analyzing the concentration degree of the phase difference distribution.

[0010] As a preferred embodiment of the elderly psychological state assessment and early warning method of the present invention, wherein: the construction of the personal reference feature set includes: Collecting EEG signals of the elderly assessment object at multiple time periods and classifying the EEG signals according to the time periods; Applying the weighted symmetric singular spectrum analysis method to the EEG signals of each time period and calculating the statistical distribution characteristics of the power spectral density ratios of each time period.

[0011] As a preferred embodiment of the elderly psychological state assessment and early warning method of the present invention, wherein: the determination of the short-term fluctuation amplitude and the cumulative deviation includes: Calculate the normalized deviation of the current power spectral density ratio from the reference feature in the corresponding period as the short-term fluctuation amplitude; Perform weighted accumulation on the short-term fluctuation amplitudes for consecutive days to obtain the cumulative deviation amount.

[0012] As a preferred solution of the elderly psychological state evaluation and early warning method described in the present invention, wherein: the determination of the abnormal degree of the psychological state includes: Compare the short-term fluctuation amplitude with a first reference value and the cumulative deviation amount with a second reference value; According to the combination of the comparison results, determine the abnormal degree of the psychological state as temporary emotional fluctuation, potential psychological problem or serious psychological state abnormality.

[0013] To further solve the above technical problems, the present invention provides the following technical solution: an elderly psychological state evaluation and early warning system, including: an electroencephalogram reference feature acquisition module, configured to collect electroencephalogram signals of an elderly evaluation object at a fixed time period every day, calculate the power spectral density ratio of the electroencephalogram signals, and construct a personal reference feature set of the elderly evaluation object; A real-time electroencephalogram signal analysis module, configured to obtain the current electroencephalogram signal of the elderly evaluation object, calculate the power spectral density ratio of the current electroencephalogram signal, compare the power spectral density ratio with the personal reference feature set, and determine the short-term fluctuation amplitude and the cumulative deviation amount; A psychological state early warning determination module, configured to determine the abnormal degree of the psychological state according to the short-term fluctuation amplitude and the cumulative deviation amount, and generate a graded early warning signal including a risk level and a timestamp.

[0014] A computer device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned elderly psychological state evaluation and early warning method are implemented.

[0015] A computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, the steps of the above-mentioned elderly psychological state evaluation and early warning method are implemented.

[0016] Advantages of the present invention: First, by introducing the weighted symmetric singular spectrum analysis method for EEG signal preprocessing and combining the inter-band phase synchronization index as a weight factor, the ability to retain weak but physiologically significant EEG signal components is improved, and the problem that traditional signal processing methods are incomplete in extracting key EEG features is solved; Second, a personal benchmark feature set is constructed to replace the traditional group standard evaluation mode, realizing personalized customization of the evaluation standard and effectively solving the technical bottleneck that it is difficult for group standards to adapt to individual differences of the elderly; Third, a dual-index evaluation system of short-term fluctuation amplitude and cumulative deviation is introduced. Through normalized deviation calculation and weighted cumulative processing, dual monitoring of sudden anomalies and progressive changes is realized, breaking through the limitation that traditional single evaluation cannot capture dynamic change trends; Fourth, a mental state abnormality determination matrix and a hierarchical early warning mechanism based on a multi-dimensional state space model are designed. By associating EEG signal feature changes with specific mental state types, accurate abnormal type recognition and risk classification early warning capabilities are provided, solving the problem that the early warning results in the prior art lack refined classification and intervention guidance. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 It is a schematic diagram of the overall process of an elderly mental state evaluation and early warning method proposed by the present invention; Figure 2 It is a computer device diagram in an elderly mental state evaluation and early warning method proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the drawings of the specification. Obviously, the described embodiments are some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0020] Many specific details are set forth in the following description in order to provide a thorough understanding of the present invention. However, the present invention may be practiced in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0021] Example 1, referring to Figure 1 , which is an embodiment of the present invention, provides a method for evaluating and warning the psychological state of the elderly.

[0022] Figure 1 The overall process schematic diagram of a method for evaluating and warning the psychological state of the elderly is shown, including the following steps: S1: Collect the electroencephalogram (EEG) signals of the elderly evaluation object at fixed time periods every day, calculate the power spectral density ratio of the EEG signals, and construct the personal benchmark feature set of the elderly evaluation object.

[0023] In this embodiment, first, it is necessary to collect the EEG signal data of the elderly evaluation object, preprocess the signal through the weighted symmetric singular spectrum analysis method, calculate the power spectral density ratio, and finally construct a personalized benchmark feature set for subsequent psychological state evaluation.

[0024] S1.1: Collect the EEG signals of the elderly evaluation object at fixed time periods every day.

[0025] In this embodiment, a portable EEG acquisition device is used to collect the EEG signals of the elderly evaluation object. The acquisition device adopts dry electrode technology, which has the characteristics of comfortable wearing and convenient operation, and is suitable for daily use by the elderly. The electrode placement positions follow the international 10-20 system, mainly including key parts such as the frontal lobe (Fp1, Fp2, F3, F4), temporal lobe (T3, T4), parietal lobe (P3, P4), and occipital lobe (O1, O2). The sampling frequency is set to 256 Hz to reduce the data storage burden while ensuring the signal quality.

[0026] To obtain reliable reference data, the collection time is selected at three fixed time periods every day for the elderly evaluation object: 9:00 - 10:00 in the morning, 15:00 - 16:00 in the afternoon, and 20:00 - 21:00 in the evening. Each collection lasts for 5 minutes. The reason for choosing these three time periods is that the elderly are usually in a relatively stable state at these time points, and they are distributed at different times of the day, which can reflect the variation law of the elderly's all-day EEG activities. During the collection process, the evaluation object is required to keep quiet and relaxed, and avoid large-scale limb movements and emotional fluctuations to reduce artifact interference.

[0027] S1.2: Calculate the power spectral density ratio of the EEG signals.

[0028] Since the original EEG signals usually contain a large amount of noise and artifacts, it is necessary to preprocess the signals and then calculate the power spectral density ratio. The specific processing steps are as follows: S1.2.1: Preprocess the EEG signals through the weighted symmetric singular spectrum analysis method.

[0029] Singular Spectrum Analysis (SSA) is an effective time - series decomposition method that can decompose a signal into components such as trends, periods, and noise. Since the traditional SSA method has inaccurate selection of signal components when dealing with electroencephalogram (EEG) signals, in this embodiment, SSA is improved by introducing weighted and symmetrized processing to better adapt to the characteristics of EEG signals.

[0030] S1.2.1.1: Embed the EEG signal to form a trajectory matrix and perform singular value decomposition on the trajectory matrix to obtain eigenvalues and eigenvectors.

[0031] First, for the collected EEG signal sequence , select an appropriate window length (usually take , and ), and construct the trajectory matrix: ; Among them, , is a subsequence with a length of .

[0032] Then, calculate the covariance matrix of the trajectory matrix, perform eigen - decomposition on to obtain eigenvalues and their corresponding eigenvectors .

[0033] Singular value decomposition can be expressed as: ; Among them, is the number of non - zero eigenvalues, is the principal component vector, is the th eigenvalue, is the th eigenvector, is the th principal component vector. This step realizes the decomposition of the original EEG signal into multiple sub - components, laying a foundation for subsequent weighted processing.

[0034] S1.2.1.2: Perform weighted and symmetrized processing on the eigenvalues and eigenvectors according to the inter - band phase synchronization index.

[0035] In traditional SSA, the principal components are usually selected according to the eigenvalue magnitudes for signal reconstruction. However, for EEG signals, some weak but physiologically significant components may be ignored. Therefore, in this embodiment, the inter - band phase synchronization index is introduced as a weighting factor to enhance the signal components with high phase synchronization and better retain the key information related to cognitive functions and emotional states in EEG signals.

[0036] First, the Hilbert transform is used to extract the instantaneous phase of each frequency band of the EEG signal. For the EEG signal Apply a band-pass filter to separate frequency band, frequency band, frequency band, and frequency band signals, denoted as , , , respectively. Then, apply the Hilbert transform to each frequency band signal to calculate its analytic signal: ; ; where, represents the frequency band, is the result of the Hilbert transform, is the analytic signal, is the instantaneous amplitude, is the instantaneous phase, is the time variable, is the integration variable, is the imaginary unit. Through the Hilbert transform, the instantaneous phase information of each frequency band signal can be extracted, which is the basis for calculating phase synchrony.

[0037] Secondly, calculate the phase difference distribution between the instantaneous phases of each frequency band and statistically analyze the concentration degree of the phase difference distribution. Calculate the phase difference between different frequency bands: ; where, represents different frequency bands, n:m is the frequency ratio, and common ratios include 1:1, 1:2, 2:1, etc., is the frequency band and the frequency band the phase difference between them. This method can capture the non-linear coupling relationship between different frequency bands and can better reflect the complex dynamic characteristics of EEG signals than traditional linear correlation analysis.

[0038] Then, use the Phase Locking Value (PLV) to quantify the degree of phase synchrony: ; where, is the total number of time points, is the frequency band and Phase-locking values between. The PLV value ranges from 0 to 1, and the closer the value is to 1, the higher the degree of phase synchronization. Research shows that the phase synchronization between different brain regions and different frequency bands is closely related to cognitive function and emotional state, especially between the frontal lobe and the parietal lobe wave and wave phase synchronization has an important impact on emotion regulation.

[0039] Based on the PLV values between each frequency band pair, construct a weight matrix , and weight the singular value decomposition result: ; Among them, is the weight factor calculated according to the PLV between frequency bands, , is the mapping function, and usually the sigmoid function is used to map the PLV value to the range of [0.5, 1.5], is the weighted signal matrix. This weighting method improves the ability to retain weak signal components, especially those weak fluctuations related to emotional state.

[0040] Finally, in order to enhance the symmetry of the signal, introduce a symmetrization operation: ; Among them, is the anti-diagonal identity matrix, is the symmetrized signal matrix. This symmetrization operation can effectively maintain the phase characteristics of the reconstructed signal, improve the signal reconstruction quality, and provide a more reliable signal basis for subsequent power spectral density calculation.

[0041] S1.2.2: Calculate the δ frequency band, θ frequency band, α frequency band and β frequency band power spectral density ratio.

[0042] Apply the fast Fourier transform FFT to the preprocessed EEG signal to calculate the power spectral density of each frequency band: ; Among them, represents the frequency range of the frequency band , is the absolute power spectral density of the frequency band .

[0043] Then calculate the relative power spectral density of each frequency band: ; Among them, For the frequency band of the relative power spectral density.

[0044] Based on the above calculation results, extract the following power spectral density ratios as characteristic indexes: : Ratio, reflecting the cognitive function state, and a decrease in this ratio is usually associated with a decline in cognitive function; : Ratio, reflecting the brain arousal level, and an increase in this ratio indicates an improvement in arousal level; : Ratio, reflecting the emotion regulation ability, and an abnormal increase in this ratio is related to the depressive state; : Ratio, reflecting the attention level, and if this ratio is too high, it indicates inattentiveness.

[0045] Compared with the traditional single-frequency band power analysis, these ratio indexes can eliminate the influence of baseline differences between individuals and more stably and reliably reflect the electroencephalogram change characteristics of the elderly assessment objects, providing a solid foundation for the construction of the subsequent personal benchmark feature set.

[0046] S1.3: Construct the personal benchmark feature set of the elderly assessment objects In order to establish a stable and reliable personal benchmark feature set, it is necessary to collect electroencephalogram data over a period of time and extract statistical features. The innovation of this step lies in establishing a personalized assessment benchmark instead of the traditional group standard, which can better adapt to the individual differences of the elderly.

[0047] S1.3.1: Collect the electroencephalogram signals of the elderly assessment objects in multiple periods and classify the electroencephalogram signals according to the periods.

[0048] In this embodiment, collect the electroencephalogram data of the elderly assessment objects for 14 consecutive days, including the data of three fixed periods (morning, afternoon, and evening) every day. The reason for setting the acquisition period to 14 days is that it can cover two complete physiological cycles of the elderly and reduce the influence of accidental factors. The collected data is sorted and classified according to the periods to form data sets for three periods. This classification method can reflect the variation laws of the physiological and psychological states of the elderly assessment objects at different times of the day and avoid the assessment deviation that may be caused by the traditional single-period acquisition.

[0049] S1.3.2: Apply the weighted symmetric singular spectrum analysis method to the electroencephalogram signals of each period and calculate the statistical distribution characteristics of the power spectral density ratio of each period.

[0050] For the EEG data of each period, apply the processing methods in S1.2.1 and S1.2.2 above to extract the power spectral density ratio. Then calculate the statistical distribution characteristics of the 14-day data, including: Mean : ; Among them, R represents a specific power spectral density ratio (such as ), is the number of days of acquisition (in this example ), is the power spectral density ratio on the th day, reflecting the average level of the power spectral density ratio; Standard deviation : ; Among them, reflecting the degree of fluctuation of the power spectral density ratio; Coefficient of variation : ; Among them, eliminates the influence of dimension and more objectively reflects the degree of dispersion of the data; Interquartile range : ; Among them, and are the first quartile and the third quartile of the power spectral density ratio respectively, reflecting the concentrated distribution range of the power spectral density ratio and being insensitive to outliers.

[0051] It should be noted that these statistical characteristics together constitute the personal benchmark feature set of the elderly assessment object and are used as the reference basis for subsequent abnormal psychological state detection. This embodiment considers both the central tendency (mean) and the degree of dispersion (standard deviation, coefficient of variation, interquartile range) in two dimensions, and can comprehensively describe the stable state and fluctuation range of the EEG characteristics of the elderly, providing a more accurate judgment basis for subsequent abnormal detection.

[0052] Compared with the traditional group standard, the personal benchmark feature set can better adapt to individual differences and improve the accuracy of abnormal psychological state detection. For example, some elderly people may be born with weaker wave activity. If evaluated by the group standard, they may be misjudged as having cognitive function decline; while the evaluation method based on the personal benchmark feature set can avoid this misjudgment.

[0053] Through the above steps, the construction of the personal benchmark feature set for the elderly assessment object is completed. This feature set reflects the distribution law of the electroencephalogram features of the assessment object in the normal state, providing a reliable reference basis for the subsequent psychological state assessment.

[0054] S2: Obtain the current electroencephalogram signal of the elderly assessment object, calculate the power spectral density ratio of the current electroencephalogram signal, compare the power spectral density ratio with the personal benchmark feature set, and determine the short-term fluctuation amplitude and the cumulative deviation amount.

[0055] Specifically, determining the short-term fluctuation amplitude and the cumulative deviation amount includes: Calculate the normalized deviation between the current power spectral density ratio and the benchmark feature of the corresponding time period as the short-term fluctuation amplitude; Perform weighted accumulation on the short-term fluctuation amplitudes of consecutive days to obtain the cumulative deviation amount.

[0056] In this embodiment, comparing the currently calculated power spectral density ratio with the personal benchmark feature set is a key step in evaluating the psychological state change of the elderly. This step is divided into two main parts: calculating the short-term fluctuation amplitude and determining the cumulative deviation amount.

[0057] The short-term fluctuation amplitude is used to quantify the deviation degree of the current psychological state of the elderly assessment object from its normal state. The calculation method is as follows: First, according to the currently collected time period (morning, afternoon, or evening), select the statistical features of the corresponding time period in the personal benchmark feature set, including the mean and the standard deviation . This method of time period matching can effectively eliminate the influence of daily physiological rhythm fluctuations, making the evaluation results more accurate and reliable.

[0058] Then, for each power spectral density ratio ( , , , ), calculate the difference between the current value and the benchmark mean respectively, and normalize it with the corresponding standard deviation to obtain the short-term fluctuation amplitude.

[0059] Specifically, let the current power spectral density ratio be , the benchmark mean of the corresponding time period be , and the standard deviation be , then the short-term fluctuation amplitude of this ratio is calculated as: ; Among them, It represents the short-term fluctuation amplitude, indicating the degree to which the current power spectral density ratio deviates from the normal state, with the unit of standard deviation. This normalization enables the comparison and comprehensive evaluation of fluctuations in different power spectral density ratios on the same scale.

[0060] For example, if the short-term fluctuation amplitude of the α / θ ratio of a certain elderly person is , while the short-term fluctuation amplitude of the δ / β ratio is , it indicates that the α / θ ratio is 2.5 standard deviations below the normal level, and the δ / β ratio is 3.0 standard deviations above the normal level, suggesting a possible risk of both cognitive decline and depressive mood.

[0061] The short-term fluctuation amplitudes of each power spectral density ratio are recorded and stored. On the one hand, they are used for current state assessment, and on the other hand, they serve as the basic data for calculating the cumulative deviation amount. Based on neurophysiological research and clinical experience, evaluation criteria for the short-term fluctuation amplitude are set. Here is an example: When , it indicates a normal state within the individual's normal fluctuation range; When , it indicates a mildly abnormal state that requires attention; When , it indicates a moderately abnormal state that requires intervention; When , it indicates a severely abnormal state that requires immediate intervention.

[0062] Meanwhile, the physiological significance of the fluctuation direction is also considered.

[0063] For example, for the α / θ ratio, a negative deviation ( ) is usually associated with cognitive decline, while a positive deviation ( ) may be associated with an excited state; for the δ / β ratio, a positive deviation ( ) is usually associated with a depressive state, while a negative deviation ( ) may be associated with an anxious state.

[0064] Furthermore, the short-term fluctuation amplitudes over consecutive days are weighted and accumulated to obtain the cumulative deviation amount.

[0065] Since the change in mental state is usually a gradual process, the fluctuations in single measurements may be affected by accidental factors and it is difficult to reflect the true change trend. Therefore, in this embodiment, the concept of cumulative deviation is introduced, and the weighted accumulation of the short-term fluctuation amplitudes for consecutive days is calculated to evaluate the long-term change trend of the mental state of the elderly evaluation object.

[0066] The specific calculation method is as follows: Save the short-term fluctuation amplitude data for the most recent consecutive 7 days. For each power spectral density ratio, calculate its cumulative deviation respectively. Assume that today is the th day, and the previous 6 days are the , ,..., th day respectively. Then the cumulative deviation of the power spectral density ratio is calculated as: ; where represents the short-term fluctuation amplitude on the th day, is the time decay weight coefficient, satisfying , and . is the cumulative deviation, representing the weighted average level of recent state changes.

[0067] It should be noted that the weight coefficient adopted in this embodiment is set as: , , , , , , . This time decay weighting method makes the influence of the most recent data on the cumulative deviation greater, which is in line with the clinical practice of mental state evaluation, that is, recent changes have a more significant impact on the current state.

[0068] The key characteristic of the cumulative deviation is that it can amplify persistent deviations and suppress random fluctuations. For example, if the α / θ ratio of the elderly evaluation object for 7 consecutive days is lower than the normal level, even if the short-term fluctuation amplitude per day is only -1.0 (mild abnormality), the cumulative deviation will be close to -1.0, showing an obvious downward trend; if the short-term fluctuation amplitudes alternate between positive and negative within 7 days, the cumulative deviation may be close to 0, indicating no obvious trend change.

[0069] Based on neurophysiological research and clinical practice, evaluation criteria are set for the cumulative deviations of different power spectral density ratios. Here is an example: When When it is [a certain value], it indicates that the long-term trend is stable and there are no obvious abnormalities; When it is [a certain value], it indicates that there are mild trend changes and close monitoring is required; When it is [a certain value], it indicates that there are obvious trend changes and early intervention is recommended; When it is [a certain value], it indicates that there are severe trend changes and immediate intervention and professional evaluation are required.

[0070] In practical applications, the short-term fluctuation amplitudes and cumulative deviation amounts of multiple power spectral density ratios can be comprehensively considered to form a multi-dimensional evaluation matrix. For example, if α / θ the cumulative deviation amount of the ratio and at the same time δ / β the cumulative deviation amount of the ratio it may indicate that the elderly evaluation object is experiencing a composite state change of cognitive function decline accompanied by depressive mood, and more comprehensive intervention measures are required.

[0071] Through the dual-index system of short-term fluctuation amplitude and cumulative deviation amount, the present invention can achieve comprehensive dynamic monitoring of the psychological state changes of the elderly, can capture sudden abnormalities, and can also detect progressive changes, providing a scientific basis for precise intervention. Compared with traditional single evaluations, this method based on the individual baseline feature set and time series analysis significantly improves the accuracy and predictability of the evaluation, can identify potential risks before the problem worsens, and realizes early warning of abnormal psychological states.

[0072] Practical applications show that the present invention has a large improvement in the prediction accuracy of common psychological state abnormalities of the elderly, and can generally detect potential problems 7 - 10 days in advance. Especially for state changes with unclear early symptoms such as mild cognitive impairment and senile depression, the sensitivity of this method is significantly better than that of traditional evaluation methods, providing technical support for preventing problems before they occur.

[0073] S3: Determine the degree of abnormal psychological state based on the short-term fluctuation amplitude and cumulative deviation amount, and generate a graded warning signal including a risk level and a timestamp.

[0074] Based on the short-term fluctuation amplitude and cumulative deviation amount calculated through S2, this step will further determine the degree of abnormal psychological state of the elderly evaluation object and generate a corresponding graded warning signal. This step is a key link in the entire evaluation and warning method, converting technical data into understandable risk levels and providing clear decision-making bases for caregivers and medical staff.

[0075] S3.1: Compare the short-term fluctuation amplitude with the first reference value and the cumulative deviation amount with the second reference value In this step, the short-term fluctuation amplitude and cumulative deviation calculated in S2 are compared with preset reference values to determine whether the mental state of the elderly assessment object is abnormal.

[0076] S3.1.1: Compare the short-term fluctuation amplitude with the first reference value.

[0077] The first reference value is a set of thresholds for evaluating the abnormality degree of the short-term fluctuation amplitude, which is determined based on a large amount of clinical data and expert consensus. In this embodiment, the first reference value can be set to include two thresholds, for example: and . By comparing with these two thresholds, the short-term fluctuation amplitude can be divided into three levels: When , it is determined to be normal or slightly fluctuating, denoted as ; When , it is determined to be moderately fluctuating, denoted as ; When , it is determined to be severely fluctuating, denoted as .

[0078] For each power spectral density ratio ( , , , ), its fluctuation level is calculated. For example, if the short-term fluctuation amplitude of the α / θ ratio of a certain elderly person is , then its fluctuation level is , indicating moderate fluctuation; if the short-term fluctuation amplitude of the δ / β ratio is , then its fluctuation level is , indicating severe fluctuation.

[0079] S3.1.2: Compare the cumulative deviation with the second reference value.

[0080] The second reference value is a set of thresholds for evaluating the abnormality degree of the cumulative deviation. In this embodiment, the second reference value can also be set to include two thresholds, for example: and . By comparing with these two thresholds, the cumulative deviation can be divided into three levels: When , it is determined to be a stable trend, denoted as ; When , it is determined to be a changing trend, denoted as ; When , it is determined as an obvious change trend and denoted as .

[0081] Similar to the short-term fluctuation amplitude, the trend level is calculated for each power spectral density ratio. For example, if the cumulative deviation of the ratio of α / θ for a certain elderly person , then its trend level is , indicating a change trend; if δ / β the cumulative deviation of the ratio , then its trend level is , indicating an obvious change trend.

[0082] Compared with the traditional method that only uses a single index or a fixed threshold, this embodiment comprehensively considers two dimensions of short-term fluctuation and long-term trend, and can more comprehensively evaluate the dynamic changes of the psychological state of the elderly. Especially through the introduction of the cumulative deviation, the present invention can effectively identify the slowly developing but persistent state changes, which is of great significance for early detection of potential mental health problems.

[0083] S3.2: According to the combination of the comparison results, determine the degree of mental state abnormality as temporary emotional fluctuation, potential psychological problem or serious mental state abnormality.

[0084] After comparing the short-term fluctuation amplitude and the cumulative deviation with their respective reference values, based on the combination of the comparison results, determine the degree of mental state abnormality of the elderly evaluation object. The innovation of this step lies in improving the accuracy and scientificity of the judgment through the multi-dimensional combination of short-term and long-term indicators.

[0085] S3.2.1: Construct a mental state abnormality determination matrix, and determine the degree of mental state abnormality according to the combination of the short-term fluctuation amplitude level and the cumulative deviation level.

[0086] First, construct a 3×3 determination matrix, with the horizontal axis being the short-term fluctuation amplitude level , and the vertical axis being the cumulative deviation level . The matrix elements are the determination results of the degree of mental state abnormality under the corresponding combination, as shown in Table 1: Table 1 Determination Matrix Table Based on the above determination matrix, further divide the degree of mental state abnormality into three categories: 1. Temporary emotional fluctuation: including two situations of "temporary fluctuation" and "acute fluctuation", characterized by obvious short-term fluctuation but small cumulative deviation, indicating that it may be a transient reaction caused by short-term stimulation; 2. Potential psychological problems: Include three situations of "mild concern" and "potential problems". The characteristics are that both the short-term fluctuation and the cumulative deviation amount are at a medium level, or one indicator is normal while the other is abnormal, indicating the existence of potential mental health risks; 3. Severe psychological state abnormalities: Include two situations of "moderate abnormality" and "severe abnormality". The characteristics are that both the short-term fluctuation and the cumulative deviation amount are at a relatively high level, indicating that there may be significant mental health problems and timely intervention is required.

[0087] It should be noted that when different power spectral density ratios correspond to different degrees of abnormality, the principle of "taking the higher rather than the lower" is adopted, that is, the highest degree of abnormality is used as the final judgment result. For example, if α / θ The ratio indicates "potential psychological problems", and δ / β The ratio indicates "severe psychological state abnormality", then the final judgment is "severe psychological state abnormality".

[0088] S3.2.2: Judge the specific psychological state category based on the type of abnormal power spectral density ratio.

[0089] In addition to determining the degree of psychological state abnormality, it is also necessary to further judge the specific psychological state category based on the type of abnormal power spectral density ratio. This judgment is based on the results of a large number of clinical studies, establishing a correspondence between the changes in different power spectral density ratios and specific psychological states: 1. When α / θ The ratio is significantly reduced ( or ), it may indicate a decline in cognitive function or a decrease in attention; 2. When ( α + β ) / ( δ + θ ) The ratio is significantly reduced ( or ), it may indicate a decrease in arousal, a state of fatigue or sleep problems; 3. When δ / β The ratio is significantly increased ( or ), it may indicate an increase in depressive mood; 4. When θ / α The ratio is significantly increased ( or ), it may indicate an increase in anxiety state.

[0090] When multiple power spectral density ratios are abnormal simultaneously, it is judged that there may be a change in the composite mental state. For example, α / θ when the ratio is significantly reduced while δ / β the ratio is significantly increased, it may indicate a decline in cognitive function accompanied by depressive mood; α / θ when the ratio is significantly reduced while θ / α the ratio is significantly increased, it may indicate a decline in cognitive function accompanied by an anxiety state.

[0091] This method of classifying mental states based on power spectral density ratios shows high accuracy in clinical verification.

[0092] S3.3: Generate a graded warning signal containing the risk level and timestamp.

[0093] Based on the determination results of the previous steps, generate a graded warning signal containing the risk level and timestamp to remind caregivers and medical staff to pay attention to the changes in the mental state of the elderly assessment object in an intuitive and clear manner.

[0094] S3.3.1: Set the risk level according to the degree of mental state abnormality According to the degree of mental state abnormality, set the corresponding risk level. The specific correspondence is as follows: 1. Normal state: Green risk level (Level 0), indicating that the mental state is normal and no special attention is required; 2. Temporary emotional fluctuation: Yellow risk level (Level 1), indicating short-term fluctuations and requiring attention and observation; 3. Potential mental problems: Orange risk level (Level 2), indicating the existence of potential mental health risks and suggesting preventive intervention measures; 4. Severe mental state abnormality: Red risk level (Level 3), indicating that there may be significant mental health problems and requiring immediate intervention and professional evaluation.

[0095] This graded warning mechanism makes the warning results simple and clear, facilitating non-professionals to understand and execute. At the same time, different risk levels correspond to different intervention suggestions, ensuring the pertinence and effectiveness of the intervention measures.

[0096] S.3.3.2: Record the risk level and the corresponding time information to generate a graded warning signal.

[0097] After determining the risk level, record the detailed time information, including the date and time of warning generation, the corresponding acquisition period, and the historical data of continuous monitoring. The generated warning signal contains the following main information: 1. Risk level: Represented by both color and numerical level (0 - 3 levels); 2. Timestamp: The warning generation time accurate to minutes, e.g., 2024 - 10 - 01 09:45:32; 3. Abnormal indicators: Show which power spectral density ratios are abnormal, as well as their short - term fluctuation amplitudes and cumulative deviation amounts; 4. Possible psychological state categories: Such as cognitive function decline, depressive mood; 5. Trend: Show the trend chart of each power spectral density ratio in the past 7 days; 6. Intervention suggestions: Give targeted suggestions according to the risk level and psychological state category.

[0098] Warning signals are presented in multiple ways, including screen push on mobile terminals, sound and light prompts, SMS notifications, etc., to ensure that caregivers and medical staff can obtain warning information in a timely manner. For high - risk warnings (orange and red levels), the system will send multi - channel reminders and require confirmation of receipt.

[0099] S3.3.3: Establish a warning signal delivery mechanism to push the classified warning signals to the caregiver terminal and the medical management platform A multi - level warning signal delivery mechanism can be established to ensure that warning information can be efficiently conveyed to relevant responsible persons: 1. Caregiver terminal: Warning signals are first pushed to the caregiver's mobile device or dedicated terminal to remind them to pay attention to the changes in the psychological state of the elderly assessment object. Different risk levels adopt different push strategies. For example, the green level may only be presented in the daily report, while the red level uses an instant pop - up reminder; 2. Medical management platform: For orange and red risk levels, the system simultaneously pushes warning signals to the medical management platform so that professional medical staff can understand the situation in a timely manner and provide necessary guidance; 3. Smart home system: In an environment equipped with smart home devices, the warning system can be linked with the smart home. For example, when an orange or red warning is issued, the room light and temperature are automatically adjusted to create a more comfortable environment; 4. Recording and tracking: The system saves all warning records, establishes a mental health file for the elderly assessment object, and supports long - term tracking and trend analysis.

[0100] In addition, the warning system also has a self - learning function, which adjusts warning thresholds and rules according to the feedback from caregivers and medical staff to improve the accuracy and practicality of warnings. For example, if a certain elderly person often shows transient EEG changes caused by special activities (such as family gatherings), the system can learn to recognize this pattern and avoid generating unnecessary warnings.

[0101] In this embodiment, by integrating the information of two dimensions, namely the short-term fluctuation amplitude and the cumulative deviation amount, a set of scientific and accurate methods for determining abnormal mental states is established, and risk information is transmitted in a timely manner through graded warning signals, providing comprehensive technical support for the mental health protection of the elderly. Compared with the traditional single-dimensional assessment, this method significantly improves the accuracy and practicality of detecting abnormal mental states, especially has obvious advantages in identifying early and minor changes, and provides a powerful tool for preventing elderly mental health problems.

[0102] Embodiment 2, an embodiment of the present invention, provides a mental state assessment and warning system for the elderly, including: An electroencephalogram (EEG) reference feature acquisition module, configured to collect the EEG signals of the elderly assessment object at a fixed time period every day, calculate the power spectral density ratio of the EEG signals, and construct a personal reference feature set of the elderly assessment object; A real-time EEG signal analysis module, configured to obtain the current EEG signals of the elderly assessment object, calculate the power spectral density ratio of the current EEG signals, compare the power spectral density ratio with the personal reference feature set, and determine the short-term fluctuation amplitude and the cumulative deviation amount; A mental state warning determination module, configured to determine the degree of abnormal mental state according to the short-term fluctuation amplitude and the cumulative deviation amount, and generate a graded warning signal including a risk level and a timestamp.

[0103] Embodiment 3, referring to Figure 2 , an embodiment of the present invention, which is different from the previous embodiment in that: if the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that makes a contribution to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc that can store program codes.

[0104] The logic and / or steps represented in the flowchart or otherwise described herein can, for example, be considered as a definitional sequence of executable instructions for implementing logical functions, which can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.

[0105] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection portion (electronic device) having one or more wirings, a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, as the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or otherwise processing as appropriate, and then storing it in a computer memory.

[0106] It should be understood that the various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), and the like.

[0107] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. An elderly psychological state assessment and early warning method, characterized in that, including: Collecting the electroencephalogram (EEG) signals of the elderly assessment object at a fixed time period every day, calculating the power spectral density ratio of the EEG signals, and constructing the personal benchmark feature set of the elderly assessment object; Obtaining the current EEG signal of the elderly assessment object, calculating the power spectral density ratio of the current EEG signal, comparing the power spectral density ratio with the personal benchmark feature set, and determining the short-term fluctuation amplitude and the cumulative deviation; Judging the abnormal degree of the mental state according to the short-term fluctuation amplitude and the cumulative deviation, and generating a graded warning signal including the risk level and the time stamp.

2. The elderly psychological state evaluation and early warning method according to claim 1, characterized in that: The calculating the power spectral density ratio of the EEG signals includes: Preprocessing the EEG signals by a weighted symmetric singular spectrum analysis method; Calculating the power spectral density ratios of the δ frequency band, θ frequency band, α frequency band, and β frequency band of the preprocessed EEG signals.

3. The elderly psychological state evaluation and early warning method according to claim 2, characterized in that: The weighted symmetric singular spectrum analysis method includes: Embedding the EEG signals to form a trajectory matrix and performing singular value decomposition on the trajectory matrix to obtain eigenvalues and eigenvectors; Performing weighted and symmetric processing on the eigenvalues and the eigenvectors according to the inter-band phase synchronization index.

4. The elderly psychological state evaluation and early warning method according to claim 3, characterized in that: The calculation of the inter-band phase synchronization index includes: Extracting the instantaneous phase of each frequency band of the EEG signals by using Hilbert transform; Calculating the phase difference distribution between the instantaneous phases of each frequency band and statistically analyzing the concentration degree of the phase difference distribution.

5. The elderly psychological state evaluation and early warning method according to claim 4, characterized in that: The constructing the personal benchmark feature set includes: Collecting the EEG signals of the elderly assessment object in multiple time periods and classifying the EEG signals according to the time periods; Applying the weighted symmetric singular spectrum analysis method to the EEG signals of each time period and calculating the statistical distribution characteristics of the power spectral density ratio of each time period.

6. The elderly psychological state evaluation and early warning method according to claim 5, characterized in that: The determining the short-term fluctuation amplitude and the cumulative deviation includes: Calculating the normalized deviation between the current power spectral density ratio and the reference feature of the corresponding time period as the short-term fluctuation amplitude; Performing weighted accumulation on the short-term fluctuation amplitudes of consecutive days to obtain the cumulative deviation.

7. The elderly psychological state assessment and early warning method according to claim 6, wherein: The judging the abnormal degree of the mental state includes: Comparing the short-term fluctuation amplitude with a first reference value and the cumulative deviation with a second reference value; According to the combination of the comparison results, determining the abnormal degree of the mental state as temporary emotional fluctuation, potential psychological problem, or serious mental state abnormality.

8. An elderly psychological state evaluation and early warning system, based on the elderly psychological state evaluation and early warning method according to any one of claims 1 to 7, characterized in that: including, an EEG benchmark feature acquisition module, configured to collect the EEG signals of the elderly assessment object at a fixed time period every day, calculate the power spectral density ratio of the EEG signals, and construct the personal benchmark feature set of the elderly assessment object; a real-time EEG signal analysis module, configured to obtain the current EEG signal of the elderly assessment object, calculate the power spectral density ratio of the current EEG signal, compare the power spectral density ratio with the personal benchmark feature set, and determine the short-term fluctuation amplitude and the cumulative deviation; a mental state warning determination module, configured to judge the abnormal degree of the mental state according to the short-term fluctuation amplitude and the cumulative deviation, and generate a graded warning signal including the risk level and the time stamp.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the elderly mental state assessment and warning method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method for evaluating and warning the psychological state of the elderly according to any one of claims 1 to 7.

Images