Synchronous continuous blood pressure signal and electroencephalogram signal coupling analysis method and device and fatigue state evaluation method

Abstract

The invention provides a synchronous continuous blood pressure signal and electroencephalogram signal coupling analysis method. The method comprises the following steps: a synchronous signal preprocessing step; the synchronous continuous blood pressure signals and the single-channel electroencephalogram signals are preprocessed, and clean continuous blood pressure signals and clean electroencephalogram signals are obtained; a continuous blood pressure signal feature detection and processing step; identifying a peak point of the blood pressure waveform of each period of the clean continuous blood pressure signal; each peak point corresponds to one systolic pressure anchor point; an evoked potential construction step; by taking each systolic pressure anchor point as a reference, intercepting a clean electroencephalogram signal with a preset time length from the clean electroencephalogram signal to form an intercepted segment; aligning and averaging all the intercepted segments to obtain systolic pressure evoked potential; a fatigue evaluation index calculation step; and analyzing the systolic pressure evoked potential by using a P300 normal form to obtain a systolic pressure evoked potential P300 response serving as a fatigue evaluation index.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of biomedical signal processing and human physiological state monitoring, and particularly relates to a synchronous continuous blood pressure signal and electroencephalogram signal coupling analysis method and device and a fatigue state evaluation method. BACKGROUND

[0002] Fatigue is a typical manifestation of the decline of human physiological function, and its accurate evaluation is of great significance for ensuring work safety and maintaining health. The current mainstream fatigue evaluation methods mainly include subjective scale evaluation and single physiological signal quantitative evaluation, but both have significant limitations.

[0003] Common subjective scales include Fatigue Severity Scale (FSS), Minimum Mental State Examination (MMSE), Profile of Mood States (POMS), etc. This kind of method completes the evaluation through the subjects filling in the questionnaire or oral report, and has the following defects and deficiencies. First, it is highly subjective, and the evaluation results are significantly affected by the emotions, cognitive biases and social expectation effects of the subjects. Second, it is poor in dynamics, and the scale evaluation is a "snapshot measurement", which cannot capture the continuous dynamic evolution process of fatigue from mild accumulation to severe outbreak, and is difficult to meet the real-time warning needs. Third, there are large individual differences, and different subjects have different subjective understandings of fatigue, resulting in a lack of unified objective reference standard for the scale results.

[0004] Common single physiological signal assessments include electrocardiogram (ECG), electroencephalogram (EEG), and photo-plethysmography (PPG), which extract features to quantify fatigue. However, due to the limited physiological information coverage of a single signal, single physiological signal assessment has the following problems: EEG can directly reflect central nervous system activity (e.g., increased (θ + α) / β, decreased θ / α or β / α indicate worsening fatigue), but these indicators only characterize central fatigue and cannot correlate the relationship between the peripheral circulatory system and fatigue; ECG indirectly correlates fatigue through heart rate variability analysis, time-domain and frequency-domain feature extraction, and nonlinear feature extraction; PPG indirectly correlates fatigue through waveform features and pulse wave conduction time, but neither can directly reflect the inhibitory state of the central nervous system and is significantly affected by factors such as exercise and emotional fluctuations, resulting in insufficient accuracy when used alone. Heartbeat evoked potentials (HEPs) analyze EEG amplitude using the QRS wave of the ECG as an anchor point. There have been some studies on their use in attention assessment, but they have not been used for fatigue assessment.

[0005] The physiological mechanism of fatigue is the result of the synergistic effect of the central nervous system and the peripheral circulatory system—central inhibition leads to a decrease in peripheral blood pressure regulation, while peripheral blood pressure fluctuations, in turn, affect central blood supply and neural activity. Current technologies lack a method for synchronously analyzing the correlation between these two systems, resulting in a single assessment dimension that cannot fully cover the physiological mechanisms of fatigue. Therefore, there is an urgent need for a new method that can integrate physiological information from multiple systems and objectively and accurately quantify fatigue status. Summary of the Invention

[0006] In view of the above problems, this application aims to propose a method for the coupled analysis of synchronous continuous blood pressure signals and electroencephalogram (EEG) signals, a device for the coupled analysis of synchronous continuous blood pressure signals and EEG signals, and a method for assessing fatigue status.

[0007] The method for coupling analysis of synchronous continuous blood pressure signals and electroencephalogram (EEG) signals in this application includes:

[0008] Synchronous signal preprocessing steps: The synchronized continuous blood pressure signal and single-channel EEG signal are preprocessed to obtain a clean continuous blood pressure signal and a clean EEG signal.

[0009] Continuous blood pressure signal feature detection and processing steps: Identify the peak points of the blood pressure waveform in each cycle of a clean continuous blood pressure signal; Each peak point corresponds to a systolic blood pressure anchor point;

[0010] Evoked potential construction step; taking each systolic pressure anchor point as a reference, clean electroencephalogram signals of a predetermined duration are intercepted from the clean electroencephalogram signals to form an intercepted segment; all the intercepted segments are aligned and averaged to obtain a systolic pressure evoked potential;

[0011] Fatigue evaluation index calculation step; the systolic pressure evoked potential is analyzed by using a P300 paradigm to obtain a systolic pressure evoked potential P300 response as a fatigue evaluation index.

[0012] The method for coupling analysis of the synchronous continuous blood pressure signal and the electroencephalogram signal of the application comprises:

[0013] Synchronous signal preprocessing step; the synchronous continuous blood pressure signal and the single-channel electroencephalogram signal are preprocessed to obtain a clean continuous blood pressure signal and a clean electroencephalogram signal;

[0014] Continuous blood pressure signal feature detection and processing step; the valley points of the blood pressure waveform of each period of the clean continuous blood pressure signal are identified; each valley point corresponds to a diastolic pressure anchor point;

[0015] Evoked potential construction step; taking each diastolic pressure anchor point as a reference, clean electroencephalogram signals of a predetermined duration are intercepted from the clean electroencephalogram signals to form an intercepted segment; all the intercepted segments are aligned and averaged to obtain a diastolic pressure evoked potential;

[0016] Fatigue evaluation index calculation step; the diastolic pressure evoked potential is analyzed by using a P300 paradigm to obtain a diastolic pressure evoked potential P300 response as a fatigue evaluation index.

[0017] The method for coupling analysis of the synchronous continuous blood pressure signal and the electroencephalogram signal of the application comprises:

[0018] Synchronous signal preprocessing step; the synchronous continuous blood pressure signal and the single-channel electroencephalogram signal are preprocessed to obtain a clean continuous blood pressure signal and a clean electroencephalogram signal;

[0019] Continuous blood pressure signal feature detection and processing step; the valley points of the blood pressure waveform of each period of the clean continuous blood pressure signal are identified; each valley point corresponds to a diastolic pressure anchor point; for any two adjacent diastolic pressure anchor points, if the diastolic pressure corresponding to the diastolic pressure anchor point at the later time sequence position is greater than the diastolic pressure corresponding to the diastolic pressure anchor point at the earlier time sequence position, the diastolic pressure anchor point at the later time sequence position is determined as a rising pressure anchor point;

[0020] Evoked potential construction step; taking each rising pressure anchor point as a reference, clean electroencephalogram signals of a predetermined duration are intercepted from the clean electroencephalogram signals to form an intercepted segment; all the intercepted segments are aligned and averaged to obtain a diastolic pressure evoked potential;

[0021] The fatigue evaluation index calculation step; the diastolic pressure evoked potential is analyzed by using P300 paradigm, and the diastolic pressure evoked potential P300 response as the fatigue evaluation index is obtained.

[0022] The application discloses a coupling analysis device for synchronous continuous blood pressure signals and electroencephalogram signals.

[0023] The synchronous signal preprocessing unit is used for preprocessing the synchronous continuous blood pressure signals and the single-channel electroencephalogram signals, so as to obtain clean continuous blood pressure signals and clean electroencephalogram signals.

[0024] The continuous blood pressure signal feature detection and processing unit is used for identifying peak points of blood pressure waveforms of each period of the clean continuous blood pressure signals; each peak point corresponds to a systolic pressure anchor point.

[0025] The evoked potential construction unit is used for taking each systolic pressure anchor point as a reference, intercepting clean electroencephalogram signals of a predetermined time length from the clean electroencephalogram signals to form an intercepted segment; and all the intercepted segments are aligned and averaged to obtain a systolic pressure evoked potential.

[0026] The fatigue evaluation index calculation unit is used for analyzing the systolic pressure evoked potential by using P300 paradigm, so as to obtain a systolic pressure evoked potential P300 response as the fatigue evaluation index.

[0027] The application discloses a coupling analysis device for synchronous continuous blood pressure signals and electroencephalogram signals.

[0028] The synchronous signal preprocessing unit is used for preprocessing the synchronous continuous blood pressure signals and the single-channel electroencephalogram signals, so as to obtain clean continuous blood pressure signals and clean electroencephalogram signals.

[0029] The continuous blood pressure signal feature detection and processing unit is used for identifying valley points of blood pressure waveforms of each period of the clean continuous blood pressure signals; each valley point corresponds to a diastolic pressure anchor point.

[0030] The evoked potential construction unit is used for taking each diastolic pressure anchor point as a reference, intercepting clean electroencephalogram signals of a predetermined time length from the clean electroencephalogram signals to form an intercepted segment; and all the intercepted segments are aligned and averaged to obtain a diastolic pressure evoked potential.

[0031] The fatigue evaluation index calculation unit is used for analyzing the systolic pressure evoked potential by using P300 paradigm, so as to obtain a systolic pressure evoked potential P300 response as the fatigue evaluation index.

[0032] The application discloses a coupling analysis device for synchronous continuous blood pressure signals and electroencephalogram signals.

[0033] A synchronous signal preprocessing unit is configured to preprocess the synchronous continuous blood pressure signal and the single-channel electroencephalogram signal to obtain a clean continuous blood pressure signal and a clean electroencephalogram signal.

[0034] A continuous blood pressure signal feature detection and processing unit is configured to identify a valley point of a blood pressure waveform of each period of the clean continuous blood pressure signal, with each valley point corresponding to a diastolic pressure anchor point; for any two adjacent diastolic pressure anchor points, if a diastolic pressure corresponding to a diastolic pressure anchor point at a later time sequence position is greater than a diastolic pressure corresponding to a diastolic pressure anchor point at an earlier time sequence position, the diastolic pressure anchor point at the later time sequence position is determined as a rising pressure anchor point.

[0035] An evoked potential construction unit is configured to take each rising pressure anchor point as a reference to intercept a clean electroencephalogram signal of a predetermined time length from the clean electroencephalogram signal to form an intercepted segment; and align and average all the intercepted segments to obtain a diastolic pressure evoked potential.

[0036] A fatigue evaluation index calculation unit is configured to analyze the diastolic pressure evoked potential by using a P300 paradigm to obtain a diastolic pressure evoked potential P300 response as a fatigue evaluation index.

[0037] The fatigue state evaluation method of the present application uses the fatigue evaluation index obtained by the foregoing method to analyze the fatigue of a subject.

[0038] Preferably, when the fatigue evaluation index of the subject is lower than a predetermined threshold, the subject is considered to be in a fatigue state.

[0039] The synchronous continuous blood pressure signal and electroencephalogram coupling analysis method, device and fatigue state evaluation method of the present application overcome the subjectivity and inaccuracy of subjective scale evaluation, break through the limitations of single physiological signal analysis, provide more comprehensive fatigue state information through multi-modal signal fusion, and further provide a stable physiological index capable of directly and objectively quantifying the blood pressure-brain coupling relationship to more accurately evaluate the fatigue degree. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 FIG. 1 is a flowchart of the synchronous continuous blood pressure signal and electroencephalogram coupling analysis method of the present application.

[0041] Figure 2 FIG. 2 is a diagram of original CBP and EEG signals and their 210-240s segment local amplification signals in an example.

[0042] Figure 3 FIG. 3 is a diagram of preprocessed CBP and EEG signals and their 210-240s segment local amplification signals in an example.

[0043] Figure 4The position detection results of the CBP signals SBP and DBP after preprocessing in the example.

[0044] Figure 5 The process diagram for obtaining the indicators SBP_P300_AM and DBP_P300_AM representing the fatigue state in the example.

[0045] Figure 6 The process diagram for obtaining the indicator DBP_AS_P300_AM representing the fatigue state in the example.

[0046] Figure 7 Spearman correlation analysis of the indicators SBP_P300_AM, DBP_P300_AM and DBP_AS_P300_AM representing the fatigue state obtained in the example.

[0047] Figure 8 Comparison chart of the discrimination ability of the indicators SBP_P300_AM, DBP_P300_AM and DBP_AS_P300_AM representing the fatigue state obtained in the example. DETAILED DESCRIPTION

[0048] Hereinafter, the present application will be described in detail with reference to the accompanying drawings.

[0049] The coupling analysis method of the synchronous continuous blood pressure signal and the electroencephalogram signal of the present application will be described in detail in combination with the accompanying drawings. Figure 1

[0050] 1. Synchronous signal preprocessing.

[0051] The continuous blood pressure (CBP) signal and the electroencephalogram (EEG) signal obtained by synchronous acquisition are processed as follows:

[0052] 1) CBP signal: remove baseline drift through a Butterworth high-pass filter with a cutoff frequency of 0.5 Hz, reduce motion artifacts using a Butterworth low-pass filter with a cutoff frequency of 5 Hz, and finally eliminate power frequency interference through a 50 Hz notch filter to obtain a clean CBP signal.

[0053] ​2) EEG signal: For single-channel EEG signal, 50Hz notch filter is used to eliminate power interference. EEG data is filtered by (0.05-100) Hz band-pass filter to remove non-physiological artifacts. Then, Daubechies db4 discrete wavelet is used for wavelet transform of EEG to obtain approximation component and detail component, and the components are processed by adaptive noise perfect empirical mode decomposition to calculate intrinsic mode function. Then, independent component analysis is used to calculate the independent components of intrinsic mode function, and then sample entropy quantification is performed. The independent components corresponding to the sample entropy values satisfying Gomez Herrero condition are regarded as artifacts and set to zero. Finally, inverse independent component analysis is performed to reconstruct new approximation component and detail component, on the basis of which clean EEG signal is obtained by removing cardiac electric field artifacts, eye movement artifacts and electromyographic artifacts.

[0054] 2. CBP signal feature point detection and processing.

[0055] Based on first-order differential analysis and adaptive threshold method, the peak point (corresponding to systolic blood pressure SBP) and valley point (corresponding to diastolic blood pressure DBP) of each cycle of blood pressure waveform are automatically identified. Specifically, the first derivative of the clean CBP signal is calculated to identify the peak slope change point (zero crossing point from positive to negative, i.e. the slope changes from rising to falling), and when no valid peak is detected for 2 seconds, the detection threshold is automatically reduced, and the lower limit of the threshold (0.4×average fluctuation range) is set to prevent over-sensitivity. Based on all detected continuous CBP signal peak points, the systolic blood pressure anchor point position sequence is obtained and the corresponding anchor point systolic blood pressure sequence , based on all detected continuous CBP signal valley points, the diastolic blood pressure anchor point position sequence is obtained and the corresponding anchor point diastolic blood pressure sequence .

[0056] 3. Construction of systolic blood pressure evoked potential (SBPEP) and diastolic blood pressure evoked potential (DBPEP).

[0057] Taking each anchor point position determined in step 2 as the reference, the EEG segment centered on the anchor point, 200ms before the anchor point to 600ms after the anchor point is extracted from the synchronously collected and pre-processed EEG signal. All extracted EEG segments are aligned and averaged. The EEG signal after averaging with as the anchor point is defined as the SBPEP signal; the EEG signal after averaging with is defined as the DBPEP signal. The signal defined as DBPEP signal is defined as DBPEP signal.

[0058] 4. Extract the core fatigue evaluation index P300 AM.

[0059] Based on the SBPEP or DBPEP signal obtained in step 3, locate the time window of 250-350 ms after the anchor point, and use the classic P300 paradigm to analyze the coupling between blood pressure and brain. Calculate the average amplitude of the EEG signal in this window, and define the average amplitude value as the systolic pressure evoked potential P300 response SBP P300 AM, diastolic pressure evoked potential P300 response DBP P300 AM. The amplitude value is the core physiological index for objective evaluation of fatigue state. The deeper the fatigue, the lower the amplitude of SBP P300 AM and DBP P300 AM, reflecting the reduction of cognitive processing resources allocated by the brain to endogenous events such as blood pressure fluctuations.

[0060] 5. Optimization index based on DBP trend.

[0061] In the systolic phase, the strong pressure wave is mainly used to push the blood forward, while in the diastolic phase, the blood pressure (mainly DBP at this time) is the main driving force for maintaining the continuous perfusion of coronary arteries and brain microvessels. In a state of fatigue, the cerebral hemodynamics changes, and the fluctuation of DBP is directly related to the efficiency and stability of cerebral blood supply during diastole. Therefore, DBPEP with DBP as the anchor point can more directly capture the neural activity regulation of the brain to cope with the changes in the basic pressure of blood supply. This regulation is weakened in fatigue, thus more clearly reflected in DBPEP.

[0062] Further, the brain's processing of changes in the body is asymmetric, and the response to challenging events usually consumes more cognitive resources and produces more significant event-related potentials (such as P300). The increase in diastolic pressure can be regarded as a positive challenge that needs to be responded to, indicating that the body is in a state of needing to prepare or maintain alertness, and the brain needs to allocate more neural resources to process such events. When fatigue leads to depletion of the overall cognitive resource pool, the performance of high-demand tasks will decrease much more than that of low-demand tasks. Therefore, the decay of the evoked potential at the DBP rise anchor point will be more severe and significant than that of the evoked potential at the DBP fall anchor point in fatigue, making it a more sensitive fatigue biomarker.

[0063] Based on the above two points, using the diastolic pressure rise anchor point as the reference is a preferred and efficient embodiment of the present application. The specific implementation is as follows:

[0064] For , the DBP value at the first shot is compared with the DBP value at the second The DBP values of the two groups are compared, wherein If the formula is satisfied The first The DBP of the two groups is defined as the diastolic pressure anchor point The position of the corresponding diastolic pressure anchor point is The operation of steps 3-4 is repeated based on the diastolic pressure anchor point, and the corresponding P300 average amplitude is calculated, which is recorded as the diastolic pressure-induced potential P300 response DBP_AS_P300_AM, which can more accurately reflect the degree of fatigue.

[0065] Examples

[0066] The method is verified based on 298 healthy subjects in different fatigue states, as shown in Figures 2-8 The CBP and EEG signals of the subjects are synchronously collected in a lying state based on the evaluation of the Fatigue Severity Scale (FSS), and then the FSS scores of the subjects are counted, the traditional EEG-based fatigue state indicators (θ + α) / β, θ / α, and β / α are calculated, and the indicators SBP_P300_AM, DBP_P300_AM, and DBP_AS_P300_AM reflecting the severity of fatigue are extracted using the method proposed in the application.

[0067] Among them, Figure 4 Among them, * represents systolic pressure and o represents diastolic pressure. Figure 5 Among them, the EEG segments and the average SBPEP and DBPEP signals obtained after the SBP and DBP anchors are used to calculate the mean values of the SBPEP and DBPEP signals in the 250-350 ms window after the anchor, that is, the indicators SBP_P300_AM and DBP_P300_AM representing the fatigue state can be obtained. Figure 6 Among them, the EEG segment and the evoked potential signal obtained after the DBP anchor are used to calculate the mean value of the evoked potential signal in the 250-350 ms window after the anchor, that is, the indicator DBP_AS_P300_AM representing the fatigue state can be obtained.

[0068] Based on the FSS scores of 298 subjects, the (θ+α) / β, θ / α, and β / α of EEG, as well as SBP_P300_AM, DBP_P300_AM, and DBP_AS_P300_AM, Spearman correlation analysis was performed. The results showed that the proposed indicators for quantifying fatigue severity, SBP_P300_AM (R=-0.123, P=0.034) and DBP_P300_AM (R=-0.1), are effective. The correlation and significance of the fatigue index DBP_AS_P300_AM (R=-0.122, P=0.036) with FSS were comparable to those of the traditional EEG-based indices θ / α (R=-0.122, P=0.036) and β / α (R=-0.137, P=0.018). However, the fatigue index DBP_AS_P300_AM, quantified based on the DBP boosting anchor point, showed the strongest correlation and significance with FSS (R=-0.158, P=0.006), outperforming fatigue indices extracted from traditional EEG analysis. Figure 7 As shown.

[0069] Based on the FSS scores of 298 subjects, subjects with FSS ≥ 27 (moderate to severe fatigue) were divided into one group and subjects with FSS < 27 (negligible fatigue that does not affect daily function) were divided into another group. ROC curves of the two groups were plotted using the SBP_P300_AM, DBP_P300_AM, and DBP_AS_P300_AM indices proposed in this invention to quantify the ability of the indices to distinguish the severity of fatigue, as shown in the figure below. The area under the curve is 0.5754 (standard error 0.0393). When the threshold is set to 0.88 μV, the sensitivity for differentiation is 81.13%. The area under the curve is 0.5947 (standard error 0.0403). When the threshold is set to 0.004 μV, the sensitivity for differentiation is 81.13%. The area under the curve is 0.6150 (standard error 0.0382). When the threshold is set to -0.022μV, the sensitivity for differentiation is 81.13%, such as... Figure 8 As shown.

[0070] Compared with the prior art, this application has the following significant advantages:

[0071] It boasts high objectivity and accuracy; based entirely on objective physiological signals, it eliminates subjective bias. By analyzing the physiological mechanism of blood pressure-brain coupling, it directly addresses the changes in the central nervous system's ability to respond to endogenous events in the cardiovascular system under fatigue, resulting in more accurate and reliable assessment results.

[0072] Multi-modal information fusion, strong anti-interference ability; innovatively combines blood pressure (cardiovascular system) and EEG (central nervous system) physiological information from two different sources. This method reduces the sensitivity to specific artifacts of a single signal, and further suppresses random noise through signal averaging technology, making the extracted physiological indicators more stable.

[0073] Innovatively introduced blood pressure evoked potential (SBPEP / DBPEP); this is also the core innovation point of the present application. Compared with traditional HEP, SBPEP / DBPEP is directly related to blood pressure, which is a physiological parameter crucial to cerebral perfusion and autonomic nervous activity, and can better reveal the neurovascular coupling mechanism related to fatigue, opening up a new analysis dimension for fatigue research.

[0074] Sensitive indicators, can be analyzed in detail; by introducing an anchor point subdivision method based on blood pressure change trend (pressure rise), multiple dimensional indicators can be obtained. This makes the present application not only able to assess the overall fatigue level, but also to detect more subtle physiological state changes, with higher sensitivity and specificity.

Claims

1. A method for coupled analysis of synchronous continuous blood pressure signals and electroencephalogram (EEG) signals, comprising: Synchronous signal preprocessing steps: The synchronized continuous blood pressure signal and single-channel EEG signal are preprocessed to obtain a clean continuous blood pressure signal and a clean EEG signal. Continuous blood pressure signal feature detection and processing steps: Identify the peak points of the blood pressure waveform in each cycle of a clean continuous blood pressure signal; Each peak point corresponds to a systolic blood pressure anchor point; Steps for constructing evoked potentials: Using each systolic pressure anchor point as a reference, extract clean EEG signals of a predetermined duration from the clean EEG signals to form a segment; Align and average all the segments to obtain systolic pressure evoked potentials; The steps for calculating fatigue evaluation indicators are as follows: The P300 paradigm is used to analyze the systolic pressure evoked potentials, and the P300 response of the systolic pressure evoked potentials is obtained as a fatigue evaluation indicator.

2. A method for coupled analysis of synchronous continuous blood pressure signals and electroencephalogram (EEG) signals, comprising: Synchronous signal preprocessing steps: The synchronized continuous blood pressure signal and single-channel EEG signal are preprocessed to obtain a clean continuous blood pressure signal and a clean EEG signal. Continuous blood pressure signal feature detection and processing steps; identifying the trough points of the blood pressure waveform in each cycle of a clean continuous blood pressure signal; each trough point corresponds to a diastolic pressure anchor point; Steps for constructing evoked potentials: Using each diastolic pressure anchor point as a reference, extract clean EEG signals of a predetermined duration from the clean EEG signals to form a segment; Align and average all the segments to obtain diastolic pressure evoked potentials; Fatigue evaluation index calculation steps: Use the P300 paradigm to analyze diastolic evoked potentials and obtain the diastolic evoked potential P300 response as a fatigue evaluation index.

3. A method for coupled analysis of synchronous continuous blood pressure signals and electroencephalogram (EEG) signals, comprising: Synchronous signal preprocessing steps: The synchronized continuous blood pressure signal and single-channel EEG signal are preprocessed to obtain a clean continuous blood pressure signal and a clean EEG signal. Continuous blood pressure signal feature detection and processing steps: Identify the trough points of the blood pressure waveform in each cycle of a clean continuous blood pressure signal; Each trough point corresponds to a diastolic pressure anchor point; For any two adjacent diastolic pressure anchor points, if the diastolic pressure corresponding to the later diastolic pressure anchor point is greater than the diastolic pressure corresponding to the earlier diastolic pressure anchor point, then the later diastolic pressure anchor point is determined to be a vasopressor anchor point. The steps for constructing evoked potentials are as follows: Using each escalating anchor point as a reference, a predetermined duration of clean EEG signal is extracted from the clean EEG signal to form a segment; All segments are aligned and averaged to obtain diastolic evoked potentials. Fatigue evaluation index calculation steps: Use the P300 paradigm to analyze diastolic evoked potentials and obtain the diastolic evoked potential P300 response as a fatigue evaluation index.

4. The method for coupled analysis of synchronous continuous blood pressure signals and electroencephalogram (EEG) signals according to any one of claims 1-3, characterized in that: In the synchronization signal preprocessing step For continuous blood pressure signals, baseline drift is removed by a Butterworth high-pass filter with a cutoff frequency of 0.5 Hz, motion artifacts are reduced by a Butterworth low-pass filter with a cutoff frequency of 5 Hz, and power frequency interference is eliminated by a 50 Hz notch filter to obtain a clean continuous blood pressure signal. For single-channel EEG signals, a 50Hz notch filter is used to eliminate power frequency interference; a bandpass filter of 0.05-100Hz is used to filter and remove non-physiological artifacts; wavelet transform is performed using Daubechies db4 discrete wavelets to obtain approximate and detail components, and then the components are processed by complete empirical mode decomposition with adaptive noise to calculate intrinsic mode functions; independent component analysis is used to calculate the independent components of the intrinsic mode functions, followed by sample entropy quantization; independent components whose sample entropy values ​​satisfy the Gomez-Herrero condition are considered artifacts and set to zero; inverse independent component analysis is performed to reconstruct new approximate and detail components, resulting in a clean EEG signal.

5. The method for coupled analysis of synchronous continuous blood pressure signals and electroencephalogram (EEG) signals according to any one of claims 1-3, characterized in that: In the continuous blood pressure signal feature detection and processing steps, the peak or trough value of the blood pressure waveform in each cycle is identified based on first-order differential analysis and adaptive thresholding method.

6. A device for coupling and analyzing synchronous continuous blood pressure signals and electroencephalogram (EEG) signals, comprising: The synchronous signal preprocessing unit is used to preprocess the synchronous continuous blood pressure signal and single-channel EEG signal to obtain a clean continuous blood pressure signal and a clean EEG signal. The continuous blood pressure signal feature detection and processing unit is used to identify the peak points of the blood pressure waveform in each cycle of a clean continuous blood pressure signal; each peak point corresponds to a systolic blood pressure anchor point. The evoked potential construction unit uses each systolic pressure anchor point as a reference to extract clean EEG signals of a predetermined duration to form a segment; all the segments are aligned and averaged to obtain systolic pressure evoked potentials. The fatigue evaluation index calculation unit uses the P300 paradigm to analyze the systolic pressure evoked potential and obtains the systolic pressure evoked potential P300 response as a fatigue evaluation index.

7. A device for coupling and analyzing synchronous continuous blood pressure signals and electroencephalogram (EEG) signals, comprising: The synchronous signal preprocessing unit is used to preprocess the synchronous continuous blood pressure signal and single-channel EEG signal to obtain a clean continuous blood pressure signal and a clean EEG signal. The continuous blood pressure signal feature detection and processing unit is used to identify the trough points of the blood pressure waveform in each cycle of a clean continuous blood pressure signal; each trough point corresponds to a diastolic pressure anchor point. The evoked potential construction unit uses each diastolic pressure anchor point as a reference to extract clean EEG signals of a predetermined duration from clean EEG signals to form a segment; all the segments are aligned and averaged to obtain diastolic pressure evoked potentials. The fatigue evaluation index calculation unit uses the P300 paradigm to analyze diastolic evoked potentials and obtains the diastolic evoked potential P300 response as a fatigue evaluation index.

8. A device for coupling and analyzing synchronous continuous blood pressure signals and electroencephalogram (EEG) signals, comprising: The synchronous signal preprocessing unit is used to preprocess the synchronous continuous blood pressure signal and single-channel EEG signal to obtain a clean continuous blood pressure signal and a clean EEG signal. The continuous blood pressure signal feature detection and processing unit is used to identify the trough points of the blood pressure waveform in each cycle of a clean continuous blood pressure signal; each trough point corresponds to a diastolic pressure anchor point; for any two adjacent diastolic pressure anchor points, if the diastolic pressure corresponding to the diastolic pressure anchor point in the later time position is greater than the diastolic pressure corresponding to the diastolic pressure anchor point in the earlier time position, then the diastolic pressure anchor point in the later time position is determined to be a vasopressor anchor point. The evoked potential construction unit uses each boosting anchor point as a reference to extract clean EEG signals of a predetermined duration from the clean EEG signals to form a segment; all the segments are aligned and averaged to obtain diastolic evoked potentials. The fatigue evaluation index calculation unit uses the P300 paradigm to analyze diastolic evoked potentials and obtains the diastolic evoked potential P300 response as a fatigue evaluation index.

9. A fatigue state assessment method, which uses the fatigue evaluation index obtained by any one of claims 1-5 to analyze the fatigue of the subject.

10. The fatigue state assessment method according to claim 9, characterized in that: When a subject's fatigue assessment index falls below a predetermined threshold, the subject is considered to be in a state of fatigue.

Images