A method and apparatus for detecting cerebral blood flow

Abstract

The application relates to the technical field of blood flow detection, and discloses a cerebral blood flow detection method and device, which comprises the following steps: acquiring detection signals corresponding to multiple brain regions of a target head; determining a to-be-processed signal in a current sliding window in the detection signals; pre-processing the to-be-processed signal to obtain a cerebral blood flow dynamic parameter time sequence corresponding to each brain region; determining a cerebral blood flow activity index, a cerebral blood flow coordination index and a left-right hemisphere symmetry index based on the cerebral blood flow dynamic parameter time sequence; the cerebral blood flow activity index is used for representing the active degree of cerebral blood flow of the whole brain, the cerebral blood flow coordination index is used for representing the cerebral blood flow coordination between brain regions, and the left-right hemisphere symmetry index is used for representing the symmetry change of blood flow perfusion and blood flow regulation between the left and right hemispheres. Through the sliding window technology and real-time signal processing, the application can realize rapid monitoring of the dynamic change of cerebral blood flow, and can comprehensively evaluate the functional state of the brain region and the mutual relationship from multiple dimensions.

Description

Technical Field

[0001] This invention relates to the field of blood flow detection technology, specifically to a method and device for detecting cerebral blood flow. Background Technology

[0002] Cerebral blood flow refers to the flow of blood to the brain. Normal blood flow is essential for maintaining brain function. Any abnormality in cerebral blood flow can lead to damage or death of nerve cells, thereby causing various neurological diseases.

[0003] Currently, techniques for measuring and monitoring cerebral blood flow mainly include magnetic resonance imaging (MRI), computed tomography (CT), and near-infrared spectroscopy (NIRS). While these techniques have made some progress in clinical applications, they still have some limitations. Current cerebral blood flow monitoring techniques primarily focus on measuring blood flow in localized areas. This limitation poses a challenge in assessing the dynamics of blood flow throughout the entire brain and its complex interrelationships with cognitive functions. Although some imaging techniques can provide overall brain images, limitations in resolution and temporal resolution make it difficult to effectively capture real-time changes in blood flow across the entire brain. Therefore, there is an urgent need to develop new methods to achieve a comprehensive assessment of whole-brain blood flow, thereby better understanding brain function and its role in various neurological diseases. Summary of the Invention

[0004] This invention provides a method and device for detecting cerebral blood flow, in order to solve the problem that it is difficult to achieve real-time comprehensive assessment of whole-brain blood flow in the prior art.

[0005] In a first aspect, the present invention provides a method for detecting cerebral blood flow, the method comprising:

[0006] Acquire detection signals corresponding to multiple brain regions of the target head;

[0007] Determine the signal to be processed within the current sliding window of the detected signal;

[0008] The signal to be processed is preprocessed to obtain the time series of cerebral blood flow dynamic parameters corresponding to each brain region;

[0009] Based on the time series of dynamic parameters of cerebral blood flow, the cerebral blood flow activity index, the cerebral blood flow coordination index, and the left and right hemisphere symmetry index were determined. Among them, the cerebral blood flow activity index is used to characterize the degree of cerebral blood flow activity in the whole brain, the cerebral blood flow coordination index is used to characterize the synergy of cerebral blood flow between brain regions, and the left and right hemisphere symmetry index is used to characterize the symmetrical changes in blood perfusion and blood flow regulation in the left and right hemispheres.

[0010] This invention enables rapid monitoring of dynamic changes in cerebral blood flow through sliding window technology and real-time signal processing. By combining cerebral blood flow activity index, cerebral blood flow coordination index, and left-right hemisphere symmetry index, it can comprehensively assess the functional status of brain regions and their interrelationships from multiple dimensions. This can help identify early signs of potential neurological diseases or injuries and promote clinical applications.

[0011] In one optional implementation, based on the time series of dynamic parameters of cerebral blood flow, the cerebral blood flow activity index, the cerebral blood flow coordination index, and the left-right hemisphere symmetry index are determined, including:

[0012] Based on the time series of dynamic parameters of cerebral blood flow, the correlation intensity of blood flow signals between two different brain regions was determined.

[0013] Short-time Fourier transforms were performed on the time series of each cerebral blood flow dynamic parameter to determine the power spectrum of each brain region in a preset frequency band.

[0014] Based on the power spectrum, the relative energy ratio of each brain region in the preset frequency band is determined;

[0015] Based on the time series of dynamic parameters of cerebral blood flow, the complexity index of each brain region is determined;

[0016] Based on blood flow signal correlation intensity, relative energy ratio, and complexity index, the cerebral blood flow activity index, cerebral blood flow coordination index, and left-right hemisphere symmetry index were determined.

[0017] In one alternative implementation, the correlation strength of blood flow signals between two different brain regions is determined based on a time series of dynamic parameters of cerebral blood flow, including:

[0018] The correlation intensity of blood flow signals is calculated based on a pre-established cross-correlation function;

[0019] The cross-correlation function is:

[0020] ;

[0021] in, Indicates the length of the segment captured within the sliding window. Representing discrete time delay, n represents the sampling point number within the sliding window, k represents the k-th sliding window, i represents the i-th brain region, and j represents the j-th brain region. , This represents the time series of dynamic parameters of cerebral blood flow in the i-th brain region. This represents the center time of the k-th sliding window. Indicates the sampling interval. This represents the discrete time series of cerebral blood flow dynamic parameters for the i-th brain region within the k-th sliding window. , Represents the time series of dynamic parameters of cerebral blood flow in the j-th brain region. This represents the discrete time series of cerebral blood flow dynamic parameters in the j-th brain region within the k-th sliding window. express Standard deviation express Standard deviation express mean difference express mean difference This represents the signals of the i-th brain region and the j-th brain region within the k-th sliding window. The intensity of blood flow signals is related to the signal strength.

[0022] In one optional implementation, based on the power spectrum, the relative energy ratio of each brain region in a preset frequency band is determined, including:

[0023] Determine the preset frequency band as Total bandwidth is ;

[0024] The preset frequency band energy is determined as follows: ;

[0025] The total bandwidth energy is determined as follows: ;

[0026] in, Indicates the first The brain region in the first Power spectrum within a sliding window;

[0027] The ratio of the preset frequency band energy to the total bandwidth energy is used as the relative energy ratio.

[0028] In one optional implementation, a cerebral blood flow activity index is determined based on a relative energy ratio and a complexity index; wherein:

[0029] Determine the relative energy ratio matrix and complexity index matrix of multiple brain regions within the current sliding window;

[0030] Determine the first mean and first standard deviation of the relative energy ratio matrix;

[0031] Determine the second mean and second standard deviation of the complexity index matrix;

[0032] Based on the relative energy ratio matrix, the first mean and the first standard deviation, the normalized relative energy ratio matrix is ​​determined.

[0033] Based on the complexity index matrix, the second mean, and the second standard deviation, the normalized complexity index matrix is ​​determined.

[0034] By performing a weighted average of the normalized relative energy ratio matrix and the normalized complexity index matrix, the cerebral blood flow activity index of multiple brain regions under the current sliding window is obtained.

[0035] In one optional implementation, a cerebral blood flow coordination index is determined based on the correlation intensity of blood flow signals; wherein, the index includes:

[0036] Determine the correlation intensity matrix of blood flow signals in multiple brain regions;

[0037] Determine the spatiotemporal weighting function;

[0038] The cerebral blood flow coordination index is determined based on the blood flow signal correlation intensity matrix and the spatiotemporal weighting function.

[0039] In one alternative implementation, the spatiotemporal weighting function is: ;

[0040] in, , ;

[0041] in, Indicates the current time, Represents time constant, Let s represent the center time of the k-th sliding window, s represent the step size, i represent the i-th brain region, and j represent the j-th brain region. This represents the correlation intensity of blood flow signals between the i-th and j-th brain regions within the k-th sliding window, under zero time lag. This represents the correlation intensity of blood flow signals between the p-th brain region and the q-th brain region within the k-th sliding window, under zero time lag.

[0042] The spatiotemporal adaptive fusion algorithm proposed in this invention achieves dynamic collaborative modeling of blood flow signals from different brain regions by introducing a time-space weighting function, and comprehensively analyzes the fluctuation intensity, phase relationship, and complexity characteristics of cerebral blood flow. Compared with traditional independent channel analysis, this algorithm can more sensitively capture functional coupling and regulatory differences between brain regions, improving the sensitivity and specificity of cerebral blood flow activity monitoring.

[0043] In one optional implementation, the symmetry index of the left and right hemispheres is determined based on the relative energy ratio; wherein, it includes:

[0044] Based on the relative energy ratio of each brain region in the preset frequency band, the average energy ratio of the left hemisphere corresponding to the left hemisphere brain region and the average energy ratio of the right hemisphere corresponding to the right hemisphere brain region are determined.

[0045] Based on the average energy ratio of the left hemisphere and the average energy ratio of the right hemisphere, the symmetry index of the left and right hemispheres is determined.

[0046] In one alternative implementation, the method further includes:

[0047] Based on the cross-correlation function, the phase lag between two different brain regions is determined;

[0048] Based on phase delay, the time delay matrix of multiple brain regions is determined;

[0049] Based on the time delay matrix, the dominant transmission pathway of cerebral blood flow is determined.

[0050] This embodiment effectively reveals the information transmission and functional connections between different brain regions by identifying the main transmission pathways of cerebral blood flow information. This method not only provides clear dynamic characteristics of cerebral blood flow but also offers important evidence for understanding brain network activity and its role in cognitive processes.

[0051] In a second aspect, the present invention provides a cerebral blood flow detection device, the device comprising:

[0052] A flexible headband structure designed to fit snugly against the target head;

[0053] Multiple detection modules, each located in the first preset area of ​​the flexible head-mounted structure;

[0054] The signal processing module is electrically connected to the detection module and is used to execute the cerebral blood flow detection method in any of the above embodiments.

[0055] The power supply module is located in the second preset area of ​​the flexible head-mounted structure and is electrically connected to the signal processing module.

[0056] The display and storage module, which communicates with the signal processing module, is used to display the cerebral blood flow activity index, the cerebral blood flow coordination index, and the left and right hemisphere symmetry index.

[0057] The cerebral blood flow detection device provided by this invention realizes the transformation from local blood flow measurement to whole brain function assessment, and achieves quantitative assessment of cerebral blood flow function status, which has important clinical value. Attached Figure Description

[0058] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0059] Figure 1 This is a schematic diagram of the first process of the cerebral blood flow detection method according to an embodiment of the present invention;

[0060] Figure 2 This is a first schematic diagram of a cerebral blood flow detection device according to an embodiment of the present invention;

[0061] Figure 3 This is a second schematic diagram of a cerebral blood flow detection device according to an embodiment of the present invention;

[0062] Figure 4 This is a structural block diagram of a cerebral blood flow detection device according to an embodiment of the present invention. Detailed Implementation

[0063] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0064] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0065] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0066] Cerebral blood flow is a core physiological parameter for maintaining neural function and metabolic activity. Current detection technologies still suffer from limitations such as insufficient resolution, sensitivity, or real-time performance. For example, diffusion-correlation spectroscopy (DCS), based on the autocorrelation analysis of scattered photon fluctuations, can be used for non-invasive detection of brain tissue microcirculation. However, existing DCS systems mainly achieve single-point or localized monitoring, relying on external optical fibers and desktop computers to calculate autocorrelation and blood flow index (BFI), resulting in complex structures, large volumes, and poor real-time performance. More importantly, traditional DCS can only obtain blood flow values, lacking quantitative assessment of the synergistic, symmetrical, and dynamic regulatory capabilities of multiple brain regions, making it difficult to meet the needs of intraoperative brain function monitoring and rehabilitation assessment. Therefore, there is an urgent need for a real-time, multi-brain-region-oriented cerebral blood flow assessment technology that achieves simultaneous detection and functional assessment during the acquisition phase, overcoming the current limitation of "only measuring, not interpreting."

[0067] In view of this, according to an embodiment of the present invention, a method for detecting cerebral blood flow is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0068] This embodiment provides a method for detecting cerebral blood flow, which can be used on servers, terminals, and mobile terminals, such as mobile phones and tablets. Figure 1 This is a flowchart of a cerebral blood flow detection method according to an embodiment of the present invention, such as... Figure 1 As shown, the process includes the following steps:

[0069] Step S101: Obtain detection signals corresponding to multiple brain regions of the target head.

[0070] The detection signals can be acquired using non-invasive techniques such as functional near-infrared spectroscopy, magnetic resonance imaging, and ultrasound, targeting multiple brain regions of the target head. The detected signals can include parameters such as cerebral blood flow, oxyhemoglobin concentration, and deoxyhemoglobin concentration. These signals are presented as time-series data, corresponding to the spatial distribution of brain regions.

[0071] In this embodiment, a detection module in a wearable cerebral blood flow detection device can be used to detect blood flow signals in each brain region separately, and the detection signals detected by each detection module can be acquired in real time through a signal detection module. The wearable cerebral blood flow detection device adopts a flexible wearable structure that can fit different areas of the head, and achieves multi-point synchronous acquisition and functional assessment of cerebral blood flow through multiple independent detection modules.

[0072] To achieve spatial distribution detection of cerebral blood flow, this embodiment divides the brain into four monitoring regions based on functional areas: the frontal lobe, parietal lobe, temporal lobe, and occipital lobe. Each region has an independent detection module, fitted to the corresponding scalp area to acquire local light signal changes. Each detection module is centrally scheduled by the signal processing module and can operate independently or synchronously, employing time-division multiplexing to avoid channel crosstalk. The device can dynamically adjust the sampling order and update frequency according to the task, reducing power consumption while ensuring signal stability. The device can also be expanded to six or eight regions (such as left and right frontal lobes, left and right temporal lobes, etc.) as needed to improve spatial resolution and assessment accuracy.

[0073] After the device is started, each detection module works simultaneously to collect the intensity changes of scattered light signals from brain tissue in real time (light intensity time series) and obtain detection signals.

[0074] Step S102: Determine the signal to be processed within the current sliding window in the detected signal.

[0075] After acquiring the detection signals from each detection module, the multi-channel light intensity time series is synchronously sampled and timestamped. The continuous time series signal is divided into multiple short time periods (i.e., sliding windows) using a sliding window method, and the data in each window is used for subsequent analysis.

[0076] Step S103: Preprocess the signal to be processed to obtain the time series of cerebral blood flow dynamic parameters corresponding to each brain region.

[0077] Autocorrelation is performed within a sliding window to generate time series of blood flow dynamic parameters for each brain region, such as BFI time series, and preliminary filtering and outlier removal are performed.

[0078] In this embodiment, the cerebral blood flow index (BFI) time series is taken as an example of blood flow dynamic parameter time series. After the multi-channel optical signal acquisition is completed, the BFI time series of each brain region are first uniformly time-aligned and quality-screened to ensure the temporal consistency and analysis stability of signals from multiple regions.

[0079] Specifically, the following steps are included:

[0080] Time synchronization and interpolation correction: All channel signals are resampled and time-aligned according to the timestamp to eliminate sampling delay between channels; if there is short-term frame loss or signal interruption, it is filled by sliding interpolation.

[0081] Outlier removal and detrending: Instantaneous outliers are identified using a sliding window mean and median deviation detection method, and motion artifacts are removed using multi-scale Hampel filtering; subsequently, high-pass filtering or polynomial fitting is performed to detrend the data to eliminate slow drift or baseline drift.

[0082] Normalization and standardization processing: The signals of each channel are normalized according to the mean of their steady-state intervals to obtain a relative fluctuation sequence, denoted as... , This ensures that signals from different brain regions are comparable under the same dimensional standard.

[0083] Feature extraction preparation: For the BFI time series of each brain region, calculate the local mean within a sliding window. Standard deviation Statistics such as rate of change and energy integral were used, and... , The original blood flow index was normalized and amplitude unified. Simultaneously, data quality was assessed using the rate of change, energy integral, and effective coverage threshold (e.g., ≥80%). For time windows meeting quality requirements, a multi-brain region blood flow dynamic matrix, after amplitude normalization and artifact removal, was output. , This serves as a stable input for subsequent spatiotemporal feature analysis. When the proportion of effective samples within a certain time window is insufficient, the evaluation metrics for that time window will not be updated temporarily to avoid evaluation jitter caused by frame drops or motion artifacts.

[0084] Unlike traditional single-channel offline analysis methods, this embodiment achieves simultaneous acquisition and parallel computation of signals from multiple brain regions. It can automatically complete signal pairing, time alignment, and noise suppression during the acquisition phase, thereby improving the stability and real-time performance of the computation.

[0085] Step S104: Based on the time series of cerebral blood flow dynamic parameters, determine the cerebral blood flow activity index, cerebral blood flow coordination index, and left-right hemisphere symmetry index; wherein, the cerebral blood flow activity index is used to characterize the degree of cerebral blood flow activity in the whole brain, the cerebral blood flow coordination index is used to characterize the synergy of cerebral blood flow between brain regions, and the left-right hemisphere symmetry index is used to characterize the symmetrical changes in blood perfusion and blood flow regulation in the left and right hemispheres.

[0086] The Blood Activity Index (BAI) can be calculated using the mean, peak value, or other statistical methods of the signal to reflect the dynamics of blood flow throughout the brain. The Cerebral Blood Flow Synchronization Index (CSI) can be calculated using correlation analysis methods, such as calculating the Pearson correlation coefficient; higher synchronization indicates that blood flow activity is synchronized between different regions. The Hemispheric Flow Symmetry Index (HFSI) can be determined through relative ratio analysis to identify the functional symmetry and differences between the two hemispheres.

[0087] In this embodiment, by using sliding window technology and real-time signal processing, rapid monitoring of dynamic changes in cerebral blood flow can be achieved. At the same time, by combining cerebral blood flow activity index, cerebral blood flow coordination index and left and right hemisphere symmetry index, the functional status of brain regions and their interrelationships can be comprehensively assessed from multiple dimensions. This can help identify early signs of potential neurological diseases or injuries and promote clinical applications.

[0088] This embodiment provides a method for detecting cerebral blood flow, which can be used on servers, terminals, and mobile terminals, such as mobile phones and tablets. The method includes the following steps:

[0089] Step S201: Obtain detection signals corresponding to multiple brain regions of the target head; for details, please refer to [link to relevant documentation]. Figure 1 Step S101 of the illustrated embodiment will not be described again here.

[0090] Step S202: Determine the signal to be processed within the current sliding window of the detected signal; for details, please refer to [link to relevant documentation]. Figure 1 Step S102 of the illustrated embodiment will not be described again here.

[0091] Step S203: Preprocess the signal to be processed to obtain the time series of cerebral blood flow dynamic parameters corresponding to each brain region; for details, please refer to [link to relevant documentation]. Figure 1 Step S103 of the illustrated embodiment will not be described again here.

[0092] Step S204: Based on the time series of cerebral blood flow dynamic parameters, determine the cerebral blood flow activity index, cerebral blood flow coordination index, and left-right hemisphere symmetry index; wherein, the cerebral blood flow activity index is used to characterize the degree of cerebral blood flow activity in the whole brain, the cerebral blood flow coordination index is used to characterize the synergy of cerebral blood flow between brain regions, and the left-right hemisphere symmetry index is used to characterize the symmetrical changes in blood perfusion and blood flow regulation in the left and right hemispheres.

[0093] Specifically, step S204 includes:

[0094] Step S2041: Based on the time series of cerebral blood flow dynamic parameters, determine the correlation intensity of blood flow signals between two different brain regions.

[0095] In some optional implementations, step S2041 includes: calculating the correlation intensity of blood flow signals based on a pre-established cross-correlation function;

[0096] The cross-correlation function is:

[0097] ;

[0098] in, Indicates the length of the segment captured within the sliding window. Representing discrete time delay, n represents the sampling point number within the sliding window, k represents the k-th sliding window, i represents the i-th brain region, and j represents the j-th brain region. , This represents the time series of dynamic parameters of cerebral blood flow in the i-th brain region. This represents the center time of the k-th sliding window. Indicates the sampling interval. This represents the discrete time series of cerebral blood flow dynamic parameters for the i-th brain region within the k-th sliding window. , Represents the time series of dynamic parameters of cerebral blood flow in the j-th brain region. This represents the discrete time series of cerebral blood flow dynamic parameters in the j-th brain region within the k-th sliding window. express Standard deviation express Standard deviation express mean difference express mean difference This represents the signals of the i-th brain region and the j-th brain region within the k-th sliding window. The intensity of blood flow signals is related to the signal strength.

[0099] In this embodiment, a preprocessed multi-brain region blood flow dynamic matrix can be used. For input, where, .

[0100] To indicate by A multi-brain region blood flow dynamic matrix composed of time series of cerebral blood flow in each brain region. Indicates the first Time series of dynamic parameters of cerebral blood flow in each brain region. In this embodiment, Preferably, the time series of the blood flow index (BFI) calculated using a speckle correlation spectroscopy algorithm is used, or a series that has been normalized / standardized. With sampling interval Perform discrete sampling, at the... The length of the cut-off within each sliding time window is The segment is denoted as:

[0101] .

[0102] Subsequent cross-correlation analysis, frequency domain energy analysis, and complexity analysis were all based on this discrete blood flow index sequence. Expand.

[0103] Then based on The cross-correlation function between different brain regions was calculated to obtain the correlation intensity and phase lag of blood flow signals in different brain regions.

[0104] In this embodiment, the corresponding actual time delay is: .

[0105] Within the preset time delay search range inside, take As a brain region The correlation strength between them;

[0106] Pick and convert to This serves as a phase lag between the two brain regions.

[0107] The above quantities are calculated for all brain region pairs within each sliding window, thus obtaining the correlation coefficient matrix. and time delay matrix .

[0108] Calculating the correlation intensity of blood flow signals can be used to characterize the synergy and functional connectivity patterns of cerebral blood flow.

[0109] Step S2042: Perform short-time Fourier transform on the time series of each cerebral blood flow dynamic parameter to determine the power spectrum of each brain region in the preset frequency band.

[0110] In this embodiment, the main frequency band energy in the range of 0.05–0.5 Hz is extracted. The 0.05–0.5 Hz frequency band covers the low-frequency oscillation components related to respiration and autonomic nervous system regulation, which can effectively characterize the rhythmicity and self-regulation ability of cerebral blood flow, while filtering out high-frequency interference such as heartbeats.

[0111] Short-time Fourier transform (STFT) was used to calculate the blood flow oscillation energy and power spectrum centroid of each brain region. For the first... The first brain region Signal within a time window Perform STFT to obtain the spectrum Thus, the power spectrum is obtained: .

[0112] Step S2043: Based on the power spectrum, determine the relative energy ratio of each brain region in the preset frequency band.

[0113] In some alternative implementations, step S2043 includes:

[0114] Determine the preset frequency band as Total bandwidth is ;

[0115] The preset frequency band energy is determined as follows: ;

[0116] The total bandwidth energy is determined as follows: ;

[0117] in, Indicates the first The brain region in the first Power spectrum within a sliding window;

[0118] The ratio of the preset frequency band energy to the total bandwidth energy is used as the relative energy ratio.

[0119] In this embodiment, a preset frequency band can be set, that is, the low-frequency oscillation bandwidth of interest is: Hz, the total bandwidth of the analysis is .

[0120] The low-frequency energy (preset frequency band energy) is then: ;

[0121] The total bandwidth energy is: ;

[0122] Thus, the relative energy ratio is obtained as follows: ;

[0123] This relative energy ratio represents the proportion of low-frequency oscillations (0.05–0.5 Hz) in the total energy.

[0124] Furthermore, within the same low-frequency band, the "power spectrum centroid" can be defined as:

[0125] ;

[0126] The centroid of this power spectrum reflects the "dominant frequency" of low-frequency blood flow oscillations in this brain region. (Relative energy ratio) The larger the center of gravity of the power spectrum Stability indicates that the higher the intensity and the clearer the rhythm of cerebral blood flow oscillations, the more active the metabolic and self-regulatory activities.

[0127] Calculating the relative energy ratio and the centroid of the power spectrum can reflect the intensity of cerebral blood flow oscillations and metabolic activity, and can also capture the characteristics of blood flow oscillations induced by neural activity regulation.

[0128] Step S2044: Based on the time series of dynamic parameters of cerebral blood flow, determine the complexity index of each brain region.

[0129] In this embodiment, complexity indices such as spectral entropy, sample entropy, or multi-scale entropy can be calculated for signals in each brain region to measure the dynamic complexity of the signal. A high complexity index indicates active blood flow regulation, while a low complexity index may correspond to insufficient local blood supply or limited regulation.

[0130] Step S2045: Based on the correlation intensity of blood flow signals, relative energy ratio, and complexity index, determine the cerebral blood flow activity index, cerebral blood flow coordination index, and left-right hemisphere symmetry index.

[0131] Based on the above dynamic correlation analysis, frequency domain energy analysis and complexity analysis, the three types of features are further normalized and weighted and fused to obtain three dimensionless functional indices.

[0132] In some optional implementations, step S2045 includes: determining a cerebral blood flow activity index based on a relative energy ratio and a complexity index; wherein:

[0133] Step a1: Determine the relative energy ratio matrix and complexity index matrix of multiple brain regions under the current sliding window.

[0134] Step a2: Determine the first mean and first standard deviation of the relative energy ratio matrix.

[0135] Step a3: Determine the second mean and second standard deviation of the complexity index matrix.

[0136] Step a4: Based on the relative energy ratio matrix, the first mean and the first standard deviation, determine the normalized relative energy ratio matrix.

[0137] Step a5: Based on the complexity index matrix, the second mean, and the second standard deviation, determine the normalized complexity index matrix.

[0138] Step a6: Perform a weighted average of the normalized relative energy ratio matrix and the normalized complexity index matrix to obtain the cerebral blood flow activity index of multiple brain regions under the current sliding window.

[0139] In this embodiment, the cerebral blood flow activity index is based on a multi-brain region power spectrum and complexity-weighted result, reflecting the overall level of cerebral blood flow activity. The specific implementation method is as follows:

[0140] For the The first time window, the first The relative energy ratios and complexity indices (such as sample entropy) of each brain region are denoted as follows: First, calculate the mean and standard deviation on the baseline or historical window. And normalize it:

[0141] ; ;

[0142] The cerebral blood flow activity index is defined as a weighted average of the characteristics of each brain region:

[0143] ;

[0144] in, Number of brain regions Weights determined by experience or calibration (e.g.) ).

[0145] The larger the BAI, the higher the overall cerebral blood flow oscillation energy and regulatory complexity, and the more active the brain function.

[0146] In some optional implementations, step S2045 includes: determining a cerebral blood flow coordination index based on the correlation intensity of blood flow signals; wherein:

[0147] Step b1: Determine the correlation intensity matrix of blood flow signals in multiple brain regions.

[0148] Step b2: Determine the spatiotemporal weighting function.

[0149] Step b3: Determine the cerebral blood flow coordination index based on the blood flow signal correlation intensity matrix and the spatiotemporal weighting function.

[0150] The cerebral blood flow coordination index is obtained by normalizing the cross-correlation matrix of multiple brain regions and reflects the level of blood flow coordination between brain regions. The specific implementation method is as follows:

[0151] The first result obtained from dynamic correlation analysis The correlation strength matrix of each sliding window First, according to historical statistical intervals... Perform linear normalization:

[0152] ;

[0153] in, This represents the range of values ​​for the relevant strength. This is then combined with the spatiotemporal weights. Defined as:

[0154] ;

[0155] The cerebral blood flow coordination index describes the average coordination of blood flow changes in multiple brain regions within the current sliding window. The larger the value, the tighter the functional connectivity and the better the coordinated regulation.

[0156] This embodiment proposes a spatiotemporal adaptive fusion algorithm based on dynamic blood flow signals from multiple brain regions, introducing a spatiotemporal weighting function. Modeling the dynamic coupling relationships of signals between brain regions. The determination of the spatiotemporal weighting function is given below. The specific steps.

[0157] Sliding window segmentation: dividing blood flow signals in multiple brain regions by length Step length Perform a sliding window to obtain the first... A time window, with the center time recorded as... .

[0158] Preprocessing within the window: for each brain region sequence Detrending, filtering, normalization, and artifact removal are performed to ensure that different brain regions are comparable under the same scale; if the proportion of valid samples is less than 80%, the index update for that window is paused.

[0159] Time-domain and frequency-domain feature extraction:

[0160] Calculate the cross-correlation function within each time window. The correlation intensity matrix is ​​obtained. With time delay matrix ;

[0161] Power spectrum obtained using STFT Calculate the relative energy ratio With the center of gravity of the power spectrum ;

[0162] Computational complexity metrics (such as spectral entropy, sample entropy, multi-scale entropy, etc.).

[0163] Construction of the spatiotemporal weighting function:

[0164] To emphasize the component that is "closer to the present in time and has stronger spatial synergy," this embodiment defines the weighting function as the product of time weight and spatial weight: ;

[0165] The time weights use an exponential decay form: ;

[0166] Spatial weights are based on zero-delay correlation strength normalization: ;

[0167] Spatial weighting allows for greater weighting of brain regions with higher synergy during the fusion process.

[0168] in, Indicates the current time, This represents a time constant, used to reduce the impact of outdated windows on the current assessment. Let s represent the center time of the k-th sliding window, s represent the step size, i represent the i-th brain region, and j represent the j-th brain region. This represents the correlation intensity of blood flow signals between the i-th and j-th brain regions within the k-th sliding window, under zero time lag. This represents the correlation intensity of blood flow signals between the p-th brain region and the q-th brain region within the k-th sliding window, under zero time lag.

[0169] The spatiotemporal adaptive fusion algorithm proposed in this invention achieves dynamic collaborative modeling of blood flow signals from different brain regions by introducing a time-space weighting function, and comprehensively analyzes the fluctuation intensity, phase relationship, and complexity characteristics of cerebral blood flow. Compared with traditional independent channel analysis, this algorithm can more sensitively capture functional coupling and regulatory differences between brain regions, improving the sensitivity and specificity of cerebral blood flow activity monitoring.

[0170] In some optional implementations, step S2045 includes: determining a symmetry index for the left and right hemispheres based on the relative energy ratio; wherein:

[0171] Step c1: Based on the relative energy ratio of each brain region in the preset frequency band, determine the average energy ratio of the left hemisphere corresponding to the left hemisphere brain region and the average energy ratio of the right hemisphere corresponding to the right hemisphere brain region.

[0172] Step c2: Based on the average energy ratio of the left hemisphere and the average energy ratio of the right hemisphere, determine the symmetry index of the left and right hemispheres.

[0173] In this embodiment, normalized difference measurements based on blood flow dynamic parameters of the left and right hemispheres can effectively reflect the symmetrical changes in perfusion and regulation between the two hemispheres. The specific implementation is as follows:

[0174] Let the left hemisphere brain regions be denoted as The right hemisphere is Calculate the average low-frequency energy ratio for the left and right hemispheres respectively:

[0175] ; ;

[0176] To obtain a dimensionless symmetry index between 0 and 1, this embodiment defines:

[0177] ;

[0178] in, To prevent small constants with a denominator of zero, the closer the HFSI is to 1, the more symmetrical the perfusion and regulation in the left and right hemispheres; a decrease in the index suggests that there may be insufficient perfusion or impaired regulation on one side.

[0179] In this embodiment, the algorithms for determining the cerebral blood flow activity index, cerebral blood flow coordination index, and left-right hemisphere symmetry index can run in real time on the embedded device, employing a sliding window mechanism for continuous updates. Updates are automatically paused when the effective signal coverage falls below 80% to avoid artifact interference. Furthermore, after smoothing and thresholding, the output results are displayed in real-time by the display module, showing the trend curves and spatial distribution maps of BAI, CSI, and HFSI for each region over time.

[0180] Furthermore, the algorithm can smooth and threshold the time-series BAI, CSI, and HFSI curves, issuing alerts when indicators abnormally rise or fall and persist for a certain period. Simultaneously, the blood flow functional status of each brain region is displayed in real-time on the terminal in the form of curves and spatial maps. Through the above-described mathematical definitions and implementation process, the spatiotemporal adaptive fusion algorithm of this invention can run in real-time on an embedded platform, completing the entire process from blood flow signal acquisition to BAI / CSI / HFSI indicator output, achieving a quantitative assessment of cerebral blood flow functional status.

[0181] The cerebral blood flow detection method provided by this invention realizes the spatiotemporal fusion assessment of cerebral blood flow, which can not only detect changes in blood flow, but also quantify the brain functional state (such as activity, symmetry, and coordination) in real time.

[0182] In some alternative implementations, the method further includes:

[0183] Based on the cross-correlation function, the phase lag between two different brain regions is determined;

[0184] Based on phase delay, the time delay matrix of multiple brain regions is determined;

[0185] Based on the time delay matrix, the dominant transmission pathway of cerebral blood flow is determined.

[0186] The correlation intensity matrix can be used to reflect the degree of synchronization of blood flow changes between brain regions, while the time delay matrix can be used to identify the dominant conduction direction (such as prefrontal cortex → occipital lobe).

[0187] The time delay matrix calculated in step S2041 Indicates the first Lead / lag relationships between blood flow signals in different brain regions within a sliding window.

[0188] This embodiment can be achieved by setting a time delay threshold. To identify the dominant transmission direction, the specific rules can be set as follows:

[0189] When the relevant strength Greater than the preset threshold ,and At that time, it was believed that brain regions Changes in leading brain regions ,Right now ;

[0190] when and At that time, it was believed that brain regions Leading brain region ,Right now ;

[0191] when or At that time, it was assumed that the changes in the two brain regions were roughly synchronous or the connection was not significant, and therefore directed edges were not established.

[0192] According to the above rules, for the matrix By performing threshold determination on all elements, the dominant transmission direction map (directed graph) within the current time window can be obtained, which is used to depict the dominant transmission path of cerebral blood flow information such as "prefrontal lobe → occipital lobe".

[0193] This embodiment effectively reveals the information transmission and functional connections between different brain regions by identifying the main transmission pathways of cerebral blood flow information. This method not only provides clear dynamic characteristics of cerebral blood flow but also offers important evidence for understanding brain network activity and its role in cognitive processes.

[0194] This embodiment also provides a cerebral blood flow detection device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0195] This embodiment provides a cerebral blood flow detection device, such as... Figure 2 and Figure 3 As shown, it includes:

[0196] Flexible headgear structure 1, used to fit the target head.

[0197] Multiple detection modules 2, each detection module is located in the first preset area of ​​the flexible head-mounted structure.

[0198] The detection module in this embodiment can employ different types of optical or photoelectric detection methods to acquire brain tissue blood flow signals. The detection module is used to collect the scattering signals of brain tissue under near-infrared light irradiation in a specific wavelength band, and to acquire real-time light intensity change data for each brain region. The detection module can be an integrated optical detection component, or it can include transmitting and receiving subunits to simplify the structure and improve stability.

[0199] In this embodiment, the light source can be a single-wavelength or dual-wavelength near-infrared emission module with a wavelength of 730–900 nm, or a miniature tunable light source or a broadband LED array, to adapt to different tissue thicknesses and absorption characteristics. In addition to flexible photodiode arrays (OPDs), flexible CMOS, InGaAs photodiode arrays, or multi-pixel SPAD arrays can be used for detection to achieve higher sensitivity or faster sampling rates. In an integrated solution, the light source and detector can be co-packaged on the same flexible substrate, and emission and detection can be alternated through time-division triggering, further reducing coupling loss and structural complexity.

[0200] The first preset region is a division of brain regions that can be adjusted according to monitoring needs. For example, it can be expanded to six or eight regions (such as the left and right frontal lobes, parietal lobes, and temporal lobes) to achieve high spatial resolution, or an adaptive partitioning strategy can be adopted to dynamically adjust the activation region based on the intensity of the acquired signal and the signal-to-noise ratio. For small-scale applications such as children or animal experiments, the structure can be reduced to two or three regions for simplification.

[0201] Reference Figure 4 As shown, in this embodiment, independent detection units are deployed in the frontal lobe, parietal lobe, temporal lobe, and occipital lobe, enabling the simultaneous acquisition of blood flow signals from multiple brain regions, thus achieving higher spatial resolution and brain region coverage. Compared to single-point detection systems, this invention can reflect the dynamic distribution and changes in cerebral blood flow across different regions, providing a foundation for whole-brain function assessment.

[0202] The signal processing module 3 is electrically connected to the detection module and is used to execute the cerebral blood flow detection method in any of the above embodiments.

[0203] The signal processing module can be located at the back or side of the headgear and employs a lightweight embedded computing unit (MCU, FPGA, or SoC). This module is responsible for synchronous sampling of multi-channel signals and real-time autocorrelation calculation, generating dynamic blood flow parameter sequences for each brain region (such as BFI time series). Based on this, a multi-brain region spatiotemporal fusion evaluation algorithm is run to analyze the dynamic correlation, frequency domain energy, and complexity characteristics between brain regions, and outputs in real-time cerebral blood flow activity index, cerebral blood flow coordination index, and left-right hemisphere symmetry index for quantitative assessment of cerebral blood flow functional status.

[0204] In addition to using sliding window autocorrelation analysis, the signal processing module can also use the following methods to extract dynamic blood flow features and perform functional assessment:

[0205] Frequency domain methods: Short-time Fourier transform (STFT), wavelet transform (CWT), or empirical mode decomposition (EMD) are used to extract blood flow oscillation energy and dominant frequency components;

[0206] Complexity analysis methods: Utilize multi-scale entropy, spectral entropy, or approximate entropy indices to assess the dynamic regulation capacity of cerebral blood flow;

[0207] Data-driven approach: Convolutional neural networks (CNN) or time-series models (LSTM) are used to learn features of blood flow signals in each brain region to predict BAI, CSI and HFSI.

[0208] Fusion strategy: The above methods can be dynamically combined or switched according to the real-time signal quality to improve robustness and algorithm adaptability.

[0209] In this embodiment, the signal processing module incorporates a multi-brain-region spatiotemporal fusion assessment algorithm. Besides generating dynamic blood flow parameters for each brain region (such as BFI), it also outputs the Brain Blood Flow Activity Index (BAI), Coordination Index (CSI), and Hemisphere Symmetry Index (HFSI). This method can transform cerebral blood flow measurement results into physiologically meaningful functional indicators, achieving a leap from "quantitative measurement" to "functional assessment," making it easier for clinicians and researchers to understand and interpret.

[0210] Furthermore, the signal processing module in this embodiment employs a sliding window incremental calculation and parallel processing mechanism, enabling real-time operation on MCUs, FPGAs, or SoCs with an evaluation latency of less than 0.5 seconds. Simultaneously, it sets an effective signal coverage threshold and an anomaly suppression mechanism, ensuring stable and reliable results even under motion or noise interference environments, meeting the real-time requirements of intraoperative and continuous monitoring scenarios.

[0211] The power supply module 4 is located in the second preset area of ​​the flexible head-mounted structure, and the power supply module is electrically connected to the signal processing module.

[0212] In this embodiment, the power supply module adopts a flexible power system, integrated with the headgear, to provide stable power to the detection and signal processing module. This module may include:

[0213] Flexible lithium battery pack: embedded in the edge of the hat or the back of the head, lightweight and evenly distributed, without affecting wearing comfort;

[0214] Power Management Unit (PMU): Implements voltage regulation, overcurrent protection, and low-power control;

[0215] Wireless charging port: Supports magnetic or inductive charging, no disassembly required;

[0216] Flexible cabling network: Using conductive fabric or FPC cabling, the power supply is electrically connected to each detection module.

[0217] This design ensures that the entire machine can run continuously for ≥6 hours, meeting the requirements for long-term cerebral blood flow monitoring.

[0218] In addition to flexible lithium batteries, the power supply module can also use solid-state batteries, thin-film batteries, or energy harvesting elements (such as thermoelectric or piezoelectric generators). Power supply methods can include wireless induction, magnetic charging, or modular battery replacement to adapt to long-term or continuous monitoring scenarios.

[0219] In this embodiment, the detection, power supply, and signal processing modules are all embedded in the headcap using flexible printed circuit boards (FPCs) or conductive fabrics, achieving lightweight and flexible fit. Furthermore, the device supports modular expansion, allowing adjustment of the detection area and number of channels to meet different experimental or clinical needs, providing excellent scalability and maintainability.

[0220] The display and storage module 5 is connected in communication with the signal processing module and is used to display the cerebral blood flow activity index, the cerebral blood flow coordination index, and the left and right hemisphere symmetry index.

[0221] The test results can be wirelessly transmitted to mobile terminals (such as tablets, mobile phones, or computers) to display real-time distribution maps of blood flow dynamic parameters, cerebral blood flow activity index (BAI), and cerebral blood flow coordination index (CSI) for each brain region. The system supports local and cloud data storage for long-term monitoring, trend analysis, and multiple comparative evaluations.

[0222] In addition to being displayed on the terminal interface, the test results can also be transmitted to the host computer or cloud via Wi-Fi, Bluetooth or 5G network to achieve remote real-time monitoring, data backup and multi-device collaborative analysis.

[0223] The cerebral blood flow distribution map can be displayed in various forms, such as two-dimensional heat maps, three-dimensional model projections, or time dynamic curves; it also supports cloud-based synchronous analysis, individualized modeling, and comparison of multiple monitoring results.

[0224] The device can also generate two-dimensional or three-dimensional cerebral blood flow distribution maps based on BFI sequences from multiple brain regions and calculated BAI, CSI, and HFSI indices. This visually displays the spatial and temporal variations in cerebral blood flow, achieving a closed-loop process of "detection-calculation-evaluation-display." Through continuous monitoring and trend analysis, it can be used to assess cerebral blood flow regulation capabilities, left-right hemisphere symmetry, and functional activity status. Users can also quantitatively track and individually analyze changes in brain function based on indicator trends at different stages.

[0225] In this embodiment, the detection results can be wirelessly transmitted to a mobile terminal or host computer, displaying real-time changes in blood flow distribution and functional indices (BAI, CSI, HFSI) across multiple brain regions. It also supports local and cloud storage, facilitating individual longitudinal trend analysis and multiple monitoring comparisons, providing a visualization tool for cerebral blood flow function research and intraoperative decision support.

[0226] The cerebral blood flow detection device provided by this invention realizes the transformation from local blood flow measurement to whole brain function assessment, and achieves quantitative assessment of cerebral blood flow function status, which has important clinical value.

[0227] The cerebral blood flow detection device provided in this embodiment of the invention can execute the cerebral blood flow detection method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the various modules and units described above are the same as in the corresponding embodiments described above, and will not be repeated here.

[0228] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and all such modifications and variations fall within the scope defined by the appended claims.

Claims

1. A method for detecting cerebral blood flow, characterized in that, The method includes: Acquire detection signals corresponding to multiple brain regions of the target head; Determine the signal to be processed within the current sliding window from the detected signals; The signal to be processed is preprocessed to obtain the time series of cerebral blood flow dynamic parameters corresponding to each brain region; Based on the time series of the cerebral blood flow dynamic parameters, a cerebral blood flow activity index, a cerebral blood flow coordination index, and a left-right hemisphere symmetry index are determined; wherein, the process includes: determining the correlation intensity of blood flow signals between two different brain regions based on the time series of the cerebral blood flow dynamic parameters; performing a short-time Fourier transform on each of the cerebral blood flow dynamic parameter time series to determine the power spectrum of each brain region in a preset frequency band; determining the relative energy ratio of each brain region in the preset frequency band based on the power spectrum; determining the complexity index of each brain region based on the time series of the cerebral blood flow dynamic parameters; and determining the cerebral blood flow activity index, the cerebral blood flow coordination index, and the left-right hemisphere symmetry index based on the blood flow signal correlation intensity, the relative energy ratio, and the complexity index. The cerebral blood flow activity index is used to characterize the degree of cerebral blood flow activity in the whole brain, the cerebral blood flow coordination index is used to characterize the cerebral blood flow coordination between the brain regions, and the left and right hemisphere symmetry index is used to characterize the symmetrical changes in blood perfusion and blood flow regulation in the left and right hemispheres. The determination of the correlation intensity of blood flow signals between two different brain regions based on the time series of the cerebral blood flow dynamic parameters includes: The correlation intensity of the blood flow signal is calculated based on a pre-established cross-correlation function; The cross-correlation function is: ； in, Indicates the length of the segment captured within the sliding window. Representing discrete time delay, n represents the sampling point number within the sliding window, k represents the k-th sliding window, i represents the i-th brain region, and j represents the j-th brain region. , This represents the time series of the cerebral blood flow dynamic parameters for the i-th brain region. This represents the center time of the k-th sliding window. Indicates the sampling interval. This represents the discrete time series of cerebral blood flow dynamic parameters for the i-th brain region within the k-th sliding window. , The time sequence of the dynamic parameters of cerebral blood flow in the j-th brain region is represented. This represents the discrete time series of cerebral blood flow dynamic parameters for the j-th brain region within the k-th sliding window. express Standard deviation express Standard deviation express mean difference express mean difference This represents the signals of the i-th brain region and the j-th brain region within the k-th sliding window. The intensity of blood flow signals is related to the signal strength.

2. The method according to claim 1, characterized in that, Determining the relative energy ratio of each brain region in the preset frequency band based on the power spectrum includes: The preset frequency band is determined as Total bandwidth is ; The preset frequency band energy is determined as follows: ; The total bandwidth energy is determined as follows: ; in, Indicates the first The brain regions mentioned in the first The power spectrum within each sliding window; The ratio of the preset frequency band energy to the total bandwidth energy is used as the relative energy ratio.

3. The method according to claim 1, characterized in that, Based on the relative energy ratio and the complexity index, the cerebral blood flow activity index is determined; wherein, it includes: Determine the relative energy ratio matrix and complexity index matrix of the multiple brain regions under the current sliding window; Determine the first mean and first standard deviation of the relative energy ratio matrix; Determine the second mean and second standard deviation of the complexity index matrix; Based on the relative energy ratio matrix, the first mean and the first standard deviation, the normalized relative energy ratio matrix is ​​determined. Based on the complexity index matrix, the second mean, and the second standard deviation, the normalized complexity index matrix is ​​determined. The normalized relative energy ratio matrix and the normalized complexity index matrix are weighted and averaged to obtain the cerebral blood flow activity index of the multiple brain regions under the current sliding window.

4. The method according to claim 1, characterized in that, Based on the correlation intensity of the blood flow signal, the cerebral blood flow coordination index is determined; wherein, it includes: Determine the correlation intensity matrix of blood flow signals in the multiple brain regions; Determine the spatiotemporal weighting function; The cerebral blood flow coordination index is determined based on the blood flow signal correlation intensity matrix and the spatiotemporal weighting function.

5. The method according to claim 4, characterized in that, The spatiotemporal weighting function is: ; in, , ; in, Indicates the current time, Represents time constant, The center time of the k-th sliding window is represented by s, the step size is represented by i, the brain region is represented by j, and the brain region is represented by j. This represents the correlation intensity of blood flow signals between the i-th and j-th brain regions within the k-th sliding window, under zero time lag. The signal intensity of blood flow signal correlation between the p-th brain region and the q-th brain region within the k-th sliding window at zero time delay is represented.

6. The method according to claim 1, characterized in that, Based on the relative energy ratio, the symmetry index of the left and right hemispheres is determined; wherein, it includes: Based on the relative energy ratio of each brain region in the preset frequency band, the average energy ratio of the left hemisphere corresponding to the left hemisphere brain region and the average energy ratio of the right hemisphere corresponding to the right hemisphere brain region are determined. The symmetry index of the left and right hemispheres is determined based on the average energy ratio of the left hemisphere and the average energy ratio of the right hemisphere.

7. The method according to claim 2, characterized in that, The method further includes: Based on the cross-correlation function, the phase lag between two different brain regions is determined; Based on the phase delay, determine the delay matrix of the multiple brain regions; Based on the time delay matrix, the dominant transmission pathway of cerebral blood flow is determined.

8. A cerebral blood flow detection device, characterized in that, The device includes: A flexible headband structure designed to fit snugly against the target head; Multiple detection modules, each of which is located in a first preset area of ​​the flexible head-mounted structure; A signal processing module, electrically connected to the detection module, is used to execute the cerebral blood flow detection method according to any one of claims 1 to 6; A power supply module is located in the second preset area of ​​the flexible head-mounted structure, and the power supply module is electrically connected to the signal processing module. The display and storage module is communicatively connected to the signal processing module and is used to display the cerebral blood flow activity index, the cerebral blood flow coordination index, and the left and right hemisphere symmetry index.

Images