Cerebral disease early warning method and system based on cerebral blood flow automatic adjusting algorithm

Abstract

The invention relates to the technical field of disease early warning assistance, in particular to a cerebral disease early warning method and system based on a cerebral blood flow automatic adjusting algorithm, and the method comprises the steps: constructing a test unit, collecting a test data set based on a physiological test task group and the test unit, carrying out the brain clinical abnormality marking of the test data set, and obtaining a marked data set; training a neural network by using the labeled data set to obtain a brain disease analysis model, obtaining real-time cerebral blood flow data, obtaining a clinical abnormal feature group according to the real-time cerebral blood flow data and the brain disease analysis model, generating an early warning analysis report according to the large language model and the clinical abnormal feature group, and uploading the early warning analysis report to a doctor diagnosis and treatment end. The intelligent degree of brain early warning can be improved, and the accuracy of brain early warning is improved.

Description

Technical Field

[0001] The present invention relates to the field of disease early warning auxiliary technology, and in particular to a brain disease early warning method and system based on a cerebral blood flow automatic regulation algorithm. Background Art

[0002] In the field of clinical medicine, brain diseases such as cerebral infarction and Alzheimer's disease have always been a major hidden danger threatening human health. These diseases often develop rapidly and have serious consequences. Early warning plays an irreplaceable and key role in improving patient prognosis and reducing disability and mortality rates. Therefore, accurate and effective brain disease warning auxiliary means are increasingly becoming a difficult problem that needs to be broken through in the medical field. Timely and accurate warnings can provide doctors with accurate reports, which can further assist doctors in making timely and accurate disease judgments.

[0003] Traditional brain disease early warning methods are mostly based on single imaging examinations (such as CT, MRI) and clinical symptom analysis. However, this traditional method relies solely on images or symptoms and cannot fully capture key physiological information such as cerebral blood flow. The generated reports require high-level physician participation, and the intelligence needs to be improved. In addition, this method has a warning lag, making it difficult to keenly detect abnormalities in the early stages of the disease when no obvious structural changes have occurred. Summary of the Invention

[0004] The present invention provides a brain disease early warning method and system based on a cerebral blood flow automatic regulation algorithm, the main purpose of which is to improve the intelligence level of brain early warning and enhance the accuracy of brain early warning.

[0005] To achieve the above objectives, the present invention provides a brain disease early warning method based on a cerebral blood flow autoregulation algorithm, comprising: Identify the test patient set and construct a cerebral blood flow detection unit, wherein the cerebral blood flow detection unit includes: a transcranial Doppler detector, an optical detector, a blood flow velocity analyzer, and a blood flow analyzer, wherein the blood flow velocity analyzer is connected to the transcranial Doppler detector, and the blood flow analyzer is connected to the optical detector; Extract test patients in sequence from the test patient set, and construct test units based on the test patients and cerebral blood flow detection units; A physiological test task group is set up, and based on the physiological test task group and the test unit, cerebral blood flow detection is performed to obtain test data, wherein the physiological test task group includes one or more physiological test tasks, and the test data includes: physiological characteristics of the test patient, blood flow velocity characteristics, and blood flow characteristics; Aggregating the test data to obtain a test data set, and labeling the test data set for brain clinical abnormalities to obtain a labeled data set; Use labeled datasets to train pre-built neural networks to obtain brain disease analysis models; receiving a brain disease warning instruction, identifying a patient to be warned based on the brain disease warning instruction, measuring cerebral blood flow of the patient to be warned, and obtaining real-time cerebral blood flow data, wherein the real-time cerebral blood flow data includes: physiological characteristics of the patient to be warned, real-time blood flow velocity characteristics, and real-time blood flow characteristics; Based on the real-time cerebral blood flow data and brain disease analysis model, brain disease analysis is performed on the patients to be warned, and a clinical abnormality feature group is obtained, wherein the clinical abnormality features in the clinical abnormality feature group are all probability values; An early warning analysis report is generated based on the pre-built large language model and clinical abnormality feature group, and the early warning analysis report is uploaded to the preset physician diagnosis and treatment terminal to complete the brain disease early warning based on the cerebral blood flow automatic regulation algorithm.

[0006] Optionally, constructing a test unit based on the test patient and the cerebral blood flow detection unit includes: The test patient is fixed using a cerebral blood flow detection unit to obtain a test unit, wherein the test unit includes: a test patient and a cerebral blood flow detection unit, the transcranial Doppler detector in the cerebral blood flow detection unit is fixed on the temporal window bone of the test patient, and the transcranial Doppler detector is connected to the blood flow velocity analyzer via a connecting line, the optical detector is fixed 2 cm above the eyebrow of the test patient, and the optical detector is connected to the blood flow analyzer via a connecting line, and an optical probe and a light source are pre-installed in the optical detector.

[0007] Optionally, the cerebral blood flow detection is performed based on the physiological test task group and the test unit to obtain test data, including: extracting a first test task from the physiological test task group; Based on the test patient in the test unit, perform a first test task, and record blood flow velocity data and blood flow volume data in the step of performing the first test task; Extracting blood flow velocity features from the blood flow velocity data to obtain a blood flow velocity feature group; Extracting blood flow features from the blood flow data to obtain a blood flow feature group; Acquiring a test patient physiological characteristic group of the test patient, wherein the test patient physiological characteristic group includes a plurality of test patient physiological characteristics; The physiological characteristic group, blood velocity characteristic group and blood flow characteristic group of the test patients are combined to obtain a single-task characteristic group; The single-task feature groups are aggregated to obtain multiple single-task feature groups, and the multiple single-task feature groups are merged to obtain test data.

[0008] Optionally, the step of recording the blood flow velocity data and the blood flow volume data in the step of performing the first test task includes: Determining an initial sampling time in the step of performing the first test task; Calculate the intermediate sampling time based on the initial sampling time and the preset sampling time interval; The light intensity autocorrelation curve between the initial sampling moment and the intermediate sampling moment is recorded by an optical detector, and the blood flow velocity is detected by a transcranial Doppler detector; Recording the intermediate sampling moment as the initial sampling moment, and returning to the step of calculating the intermediate sampling moment based on the initial sampling moment and the preset sampling time interval, until the intermediate sampling moment is not less than the preset end sampling moment; The blood flow velocity and light intensity autocorrelation curves are respectively summarized to obtain a blood flow velocity group and a light intensity autocorrelation curve group, and the blood flow velocity group and the light intensity autocorrelation curve group are respectively recorded as blood flow velocity data and blood flow volume data.

[0009] Optionally, extracting blood flow velocity features from the blood flow velocity data to obtain a blood flow velocity feature group includes: Obtain the resting blood flow rate group of the trial patients; According to the preset response time, the blood flow velocity data is divided into instantaneous blood flow velocity groups, the instantaneous blood flow velocities are sequentially extracted from the instantaneous blood flow velocity groups, and the average blood flow velocity of the static blood flow velocity groups is calculated; The velocity change rate is calculated based on the average blood flow velocity and the instantaneous blood flow velocity, where the velocity change rate is expressed as: ; in, represents the rate of change of speed, represents the instantaneous blood flow velocity, represents the mean blood flow velocity; Summarizing the velocity change rates corresponding to each instantaneous blood flow velocity in the instantaneous blood flow velocity group to obtain a velocity change rate group, and calculating the brain response value based on the velocity change rate group; Confirming the maximum blood flow velocity and the minimum blood flow velocity in the bleeding flow velocity data, and determining the minimum velocity time and the maximum velocity time corresponding to the maximum blood flow velocity and the minimum blood flow velocity, respectively; The brain response value, maximum blood flow velocity, minimum blood flow velocity, minimum velocity time and maximum velocity time are combined to obtain a blood flow velocity feature group.

[0010] Optionally, calculating the brain response value based on the speed change rate group includes: The brain response value was calculated using the following formula: ; in, Represents the brain response value, Indicates the number of speed change rates in the speed change rate group, Indicates the speed change rate group The rate of change of speed, Indicates the speed change rate group The rate of change of speed, Indicates the sampling time interval.

[0011] Optionally, extracting blood flow characteristics from the blood flow data to obtain a blood flow characteristic group includes: Identify the test sampling period group, and extract the test sampling periods in the test sampling period group in sequence; determining a test light intensity curve corresponding to the test sampling period in the blood flow data, and converting the test light intensity curve into an electric field autocorrelation curve; Discretizing the electric field autocorrelation curve to obtain a fitting data point group, wherein the fitting data point group includes a plurality of fitting data points; The data is fitted using the fitting data point group to obtain the blood flow; Summarizing the blood flow corresponding to each experimental sampling period in the experimental sampling period group to obtain a blood flow sequence; Calculate the average blood flow of the blood flow series; Identifying a maximum blood flow rate and a minimum blood flow rate in a blood flow sequence, and determining a maximum flow time and a minimum flow time corresponding to the maximum blood flow rate and the minimum blood flow rate, respectively; The average blood flow, maximum blood flow, minimum blood flow, maximum flow time and minimum flow time are combined to obtain a blood flow feature group.

[0012] Optionally, performing data fitting using the fitting data point group to obtain the blood flow includes: Constructing a multi-order blood flow fitting formula, wherein the multi-order blood flow fitting formula includes factors to be fitted, and the factors to be fitted include blood flow; Using a preset fitting algorithm and a multi-order blood flow fitting formula, data fitting is performed on the factor to be fitted to obtain a fitting factor; Based on the fitting factors, the blood flow is calculated.

[0013] Optionally, the step of labeling the test dataset for clinical brain abnormalities to obtain a labeled dataset includes: Extracting test data sequentially from the test data set, determining a target patient corresponding to the test data, and obtaining a patient clinical indicator group of the target patient; Based on the patient clinical indicator group, performing brain disease analysis on the target patient to obtain a brain disease category group, wherein the brain disease category group includes one or more brain disease categories; Constructing a total disease category vector, wherein each vector element in the total disease category vector represents a brain disease; Use the brain disease category group to numerically label the total disease category vector to obtain the disease label vector; Supplementing the disease marker vector to the test data to obtain marker data; Aggregate the labeled data to obtain a labeled dataset.

[0014] To achieve the above objectives, the present invention further provides a brain disease early warning system based on a cerebral blood flow autoregulation algorithm, comprising: A test unit construction module is used to identify a test patient set and construct a cerebral blood flow detection unit, wherein the cerebral blood flow detection unit includes: a transcranial Doppler detector, an optical detector, a blood flow velocity analyzer, and a blood flow analyzer, and the blood flow velocity analyzer is connected to the transcranial Doppler detector, and the blood flow analyzer is connected to the optical detector. Test patients are sequentially extracted from the test patient set, and the test unit is constructed based on the test patients and the cerebral blood flow detection unit; a test data labeling module, configured to set a physiological test task group, and perform cerebral blood flow detection based on the physiological test task group and the test unit to obtain test data, wherein the physiological test task group includes one or more physiological test tasks, and the test data includes: physiological characteristics, blood flow velocity characteristics, and blood flow characteristics of the test patient; summarize the test data to obtain a test data set; and label the test data set for brain clinical abnormalities to obtain a labeled data set; The patient data acquisition module is used to train a pre-built neural network using a labeled data set to obtain a brain disease analysis model, receive brain disease warning instructions, identify patients to be warned based on the brain disease warning instructions, measure cerebral blood flow in the patients to be warned, and obtain real-time cerebral blood flow data, wherein the real-time cerebral blood flow data includes: physiological characteristics of the patients to be warned, real-time blood flow velocity characteristics, and real-time blood flow characteristics; The early warning report generation module is used to analyze brain diseases of patients under warning based on real-time cerebral blood flow data and brain disease analysis models, and obtain a clinical abnormality feature group, where the clinical abnormal features in the clinical abnormality feature group are all probability values. The early warning analysis report is generated based on the pre-built large language model and the clinical abnormality feature group, and the early warning analysis report is uploaded to the preset physician diagnosis and treatment terminal.

[0015] In order to solve the above problem, the present invention further provides an electronic device, comprising: a memory storing at least one instruction; and The processor executes the instructions stored in the memory to implement the above-mentioned brain disease early warning method based on the cerebral blood flow automatic regulation algorithm.

[0016] In order to solve the above problems, the present invention also provides a computer-readable storage medium, which stores at least one instruction, and the at least one instruction is executed by a processor in an electronic device to implement the above-mentioned brain disease early warning method based on the cerebral blood flow automatic regulation algorithm.

[0017] In order to solve the problems described in the background technology, the present invention first constructs a cerebral blood flow detection unit, which realizes comprehensive monitoring of cerebral blood flow velocity and flow, and provides a hardware foundation for subsequent acquisition of accurate cerebral blood flow characteristic data. A physiological test task group is set, and based on the physiological test task group and the test unit, cerebral blood flow detection is performed to obtain test data. This step simulates the changes in cerebral blood flow under different physiological states by setting a variety of physiological test tasks, and can more comprehensively collect cerebral blood flow characteristic data of test patients under various conditions, including physiological characteristics, blood flow velocity characteristics and blood flow characteristics, enriching the data dimension, and providing more sufficient and diverse data support for the subsequent establishment of an accurate brain disease analysis model, which helps to improve the model's ability to recognize different brain disease characteristics. Then, the test data set is labeled for brain clinical abnormalities to obtain a labeled data set. This step associates brain clinical abnormalities with corresponding data, providing a clearly labeled data set for subsequent neural network training. The training data set enables the neural network to learn the mapping relationship between cerebral blood flow characteristics and brain diseases. Then, by using the labeled data set to perform supervised training on the neural network, the model can learn the association between cerebral blood flow characteristics and brain disease categories, thereby constructing a model with brain disease identification and analysis capabilities, providing an intelligent analysis tool for subsequent brain disease risk assessment of patients to be warned. Furthermore, by inputting real-time cerebral blood flow data into the brain disease analysis model, the risk probability of various brain diseases in patients to be warned can be quickly and accurately analyzed, and the clinical abnormality feature group is intuitively presented in the form of probability values, providing a quantitative risk assessment for clinicians. Finally, the large language model is used to convert the complex clinical abnormality feature probability value into an easy-to-understand warning analysis report, and it is promptly transmitted to the physician, achieving efficient information transmission and visual presentation, facilitating the physician to quickly understand the patient's brain disease risk situation and assisting the physician in making decisions. Therefore, the present invention can improve the intelligence level of brain warning and enhance the accuracy of brain warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic flow chart of a brain disease early warning method based on a cerebral blood flow autoregulation algorithm provided in one embodiment of the present invention; Figure 2 This is a functional module diagram of a brain disease early warning system based on a cerebral blood flow autoregulation algorithm provided by one embodiment of the present invention; Figure 3 A schematic structural diagram of an electronic device for implementing the brain disease early warning method based on the cerebral blood flow autoregulation algorithm provided in one embodiment of the present invention.

[0019] Description of reference numerals: 1. Electronic device; 10. Processor; 11. Memory; 12. Bus.

[0020] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0021] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0022] The embodiment of the present application provides a brain disease early warning method based on the cerebral blood flow automatic regulation algorithm. The execution subject of the brain disease early warning method based on the cerebral blood flow automatic regulation algorithm includes but is not limited to at least one of the electronic devices such as the server, the terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the brain disease early warning method based on the cerebral blood flow automatic regulation algorithm can be executed by software or hardware installed on the terminal device or the server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0023] Reference Figure 1 FIG. 1 is a flow chart of a method for early warning of brain diseases based on a cerebral blood flow autoregulation algorithm according to an embodiment of the present invention. In this embodiment, the method for early warning of brain diseases based on a cerebral blood flow autoregulation algorithm includes: S1. Identify the test patient set and construct a cerebral blood flow detection unit, wherein the cerebral blood flow detection unit includes: a transcranial Doppler detector, an optical detector, a blood flow velocity analyzer and a blood flow analyzer, and the blood flow velocity analyzer is connected to the transcranial Doppler detector, and the blood flow analyzer is connected to the optical detector.

[0024] It will be understood that the test patient set refers to a collection of multiple test patients, where test patients refer to the subjects used for the test. To obtain diverse test data, the test patients include both those with brain diseases and those without brain diseases. The transcranial Doppler detector is used to monitor aortic blood flow velocity in real time. The optical detector is a diffuse correlation spectroscopy (DCS) device that integrates a near-infrared light source and a single-photon counting detector. The optical detector calculates blood flow by detecting the intensity autocorrelation function of scalp scattered light.

[0025] Furthermore, the blood velocity analyzer refers to a device with a built-in blood flow velocity analysis algorithm. This blood flow velocity analyzer can use the data collected by the transcranial Doppler detector and the blood flow velocity analysis algorithm to obtain the blood flow characteristics of the test patient. The blood flow analyzer refers to a device with a built-in blood flow analysis algorithm. This blood flow analyzer can use the data collected by the optical detector and the blood flow analysis algorithm to obtain the blood flow characteristics of the test patient.

[0026] S2. Extract test patients in sequence from the test patient set, and construct test units based on the test patients and cerebral blood flow detection units.

[0027] It can be understood that the test unit refers to a combination of a test patient capable of performing a specific test and a cerebral blood flow detection unit.

[0028] In detail, the test unit is constructed based on the test patient and the cerebral blood flow detection unit, including: The test patient is fixed using a cerebral blood flow detection unit to obtain a test unit, wherein the test unit includes: a test patient and a cerebral blood flow detection unit, the transcranial Doppler detector in the cerebral blood flow detection unit is fixed on the temporal window bone of the test patient, and the transcranial Doppler detector is connected to the blood flow velocity analyzer via a connecting line, the optical detector is fixed 2 cm above the eyebrow of the test patient, and the optical detector is connected to the blood flow analyzer via a connecting line, and an optical probe and a light source are pre-installed in the optical detector.

[0029] It should be explained that an optical probe and a light source are installed in the optical detector, wherein the optical probe refers to a contact probe containing a multimode optical fiber, and the light source refers to a long coherence laser with a wavelength of 785nm.

[0030] S3. Set up a physiological test task group, and perform cerebral blood flow detection based on the physiological test task group and the test unit to obtain test data, wherein the physiological test task group includes one or more physiological test tasks, and the test data includes: physiological characteristics of the test patient, blood flow velocity characteristics, and blood flow characteristics.

[0031] It is clear that the physiological test task refers to a task performed by a test patient and thereby detecting the cerebral blood flow regulation parameters of the test patient. The physiological test task is set by a relevant physician.

[0032] For example, a physiological test task group includes two physiological test tasks. The first physiological test task requires the patient to hold their breath during the first physiological test task (e.g., 30 seconds). During this physiological test task, the cerebral blood flow detection unit in the test unit detects the patient's cerebral blood flow parameters. The second physiological test task requires the patient to construct a phrase based on these simple words, such as "red," "yellow," "green," and "blue," based on these simple words. For example, "flame is red," the cerebral blood flow detection unit in the test unit detects the patient's cerebral blood flow parameters during this period.

[0033] It should be explained that the physiological characteristics of the test patients refer to physiological characteristics that describe the test patients, such as the test patients' age, weight, height, etc. The blood flow velocity characteristics refer to numerical values ​​that can quantify the cerebral blood flow velocity characteristics of the test patients when performing physiological test tasks. The blood flow characteristics refer to numerical values ​​that quantify the cerebral blood flow characteristics of the test patients when performing physiological test tasks. The blood flow velocity characteristics and blood flow characteristics will be described in detail later.

[0034] Specifically, the physiological test task group and the test unit are used to perform cerebral blood flow detection and obtain test data, including: extracting a first test task from the physiological test task group; Based on the test patient in the test unit, perform a first test task, and record blood flow velocity data and blood flow volume data in the step of performing the first test task; Extracting blood flow velocity features from the blood flow velocity data to obtain a blood flow velocity feature group; Extracting blood flow features from the blood flow data to obtain a blood flow feature group; Acquiring a test patient physiological characteristic group of the test patient, wherein the test patient physiological characteristic group includes a plurality of test patient physiological characteristics; The physiological characteristic group, blood velocity characteristic group and blood flow characteristic group of the test patients are combined to obtain a single-task characteristic group; The single-task feature groups are aggregated to obtain multiple single-task feature groups, and the multiple single-task feature groups are merged to obtain test data.

[0035] It is understandable that the first test task refers to the physiological test task ranked first in the physiological test task group. The blood flow velocity data refers to the data collected by the transcranial Doppler detector in the cerebral blood flow detection unit. The blood flow data refers to the data collected by the optical detector in the cerebral blood flow detection unit. The blood flow velocity feature group and the blood flow feature group refer to a combination of multiple blood flow velocity features and a combination of multiple blood flow features, respectively. The test patient physiological feature group refers to a combination of multiple test patient physiological features. The merging refers to putting multiple data into the same array, for example: merging the test patient physiological feature group, blood flow velocity feature group and blood flow feature group to obtain a single task feature group means: putting the test patient physiological feature group, blood flow velocity feature group and blood flow feature group into the same array, and the array after putting them in is the single task feature group.

[0036] Furthermore, the step of extracting blood flow velocity features from the blood flow velocity data to obtain a blood flow velocity feature group is performed by a blood flow velocity analyzer, and the step of extracting blood flow features from the blood flow data to obtain a blood flow feature group is performed by a blood flow analyzer.

[0037] In detail, the step of recording the blood flow velocity data and blood flow volume data in the step of performing the first test task includes: Determining an initial sampling time in the step of performing the first test task; Calculate the intermediate sampling time based on the initial sampling time and the preset sampling time interval; The light intensity autocorrelation curve between the initial sampling moment and the intermediate sampling moment is recorded by an optical detector, and the blood flow velocity is detected by a transcranial Doppler detector; Recording the intermediate sampling moment as the initial sampling moment, and returning to the step of calculating the intermediate sampling moment based on the initial sampling moment and the preset sampling time interval, until the intermediate sampling moment is not less than the preset end sampling moment; The blood flow velocity and light intensity autocorrelation curves are respectively summarized to obtain a blood flow velocity group and a light intensity autocorrelation curve group, and the blood flow velocity group and the light intensity autocorrelation curve group are respectively recorded as blood flow velocity data and blood flow volume data.

[0038] It can be understood that the initial sampling moment refers to the moment when the first test task begins. The sampling time interval refers to the time interval between two samples, and the sampling time interval is set manually. The intermediate sampling moment refers to the moment corresponding to the sampling time interval after the initial sampling moment. The light intensity autocorrelation curve refers to the function of the scattered light intensity between the initial sampling moment and the intermediate sampling moment with the time delay. The acquisition method of the above-mentioned light intensity autocorrelation curve is a common operation in the laser field and will not be repeated here. The blood flow velocity refers to the average blood flow velocity of the middle cerebral artery between the initial sampling moment and the intermediate sampling moment. The end sampling moment refers to the moment when the first test task is stopped.

[0039] Specifically, the blood flow velocity feature extraction is performed on the blood flow velocity data to obtain a blood flow velocity feature group, including: Obtain the resting blood flow rate group of the trial patients; According to the preset response time, the blood flow velocity data is divided into instantaneous blood flow velocity groups, the instantaneous blood flow velocities are sequentially extracted from the instantaneous blood flow velocity groups, and the average blood flow velocity of the static blood flow velocity groups is calculated; The velocity change rate is calculated based on the average blood flow velocity and the instantaneous blood flow velocity, where the velocity change rate is expressed as: ; in, represents the rate of change of speed, represents the instantaneous blood flow velocity, represents the mean blood flow velocity; Summarizing the velocity change rates corresponding to each instantaneous blood flow velocity in the instantaneous blood flow velocity group to obtain a velocity change rate group, and calculating the brain response value based on the velocity change rate group; Confirming the maximum blood flow velocity and the minimum blood flow velocity in the bleeding flow velocity data, and determining the minimum velocity time and the maximum velocity time corresponding to the maximum blood flow velocity and the minimum blood flow velocity, respectively; The brain response value, maximum blood flow velocity, minimum blood flow velocity, minimum velocity time and maximum velocity time are combined to obtain a blood flow velocity feature group.

[0040] It is clear that the resting blood flow rate group refers to a combination of multiple cerebral blood flows of the test patient without performing any physiological test tasks. The response time refers to an artificially set duration, which indicates the duration of time for the brain to respond to the physiological test task, for example: 30 seconds. The instantaneous blood flow rate group refers to a combination of multiple instantaneous blood flow rates, wherein the instantaneous blood flow rate refers to the blood flow rate whose acquisition time is within the response time. The average blood flow rate refers to the average value of all the resting blood flow rates in the resting blood flow rate group. The velocity change rate refers to a numerical value that quantifies the relative amplitude of the deviation of the instantaneous blood flow rate from the average blood flow rate. The larger the velocity change value, the greater the blood flow velocity fluctuation caused by the stimulation of the physiological test task. The brain reaction value refers to a numerical value that quantifies the average change rate of the blood flow velocity within the response time. The larger the brain reaction value, the faster the brain reacts to the physiological test task.

[0041] Furthermore, the maximum blood flow velocity and minimum blood flow velocity refer to the maximum and minimum blood flow velocity values ​​in the blood flow velocity data, respectively. These maximum and minimum blood flow velocities can represent the extreme values ​​of blood flow velocity during the physiological test task (reflecting the limit of cerebral vascular regulation). Therefore, the maximum and minimum blood flow velocities are both considered as blood flow velocity features. The minimum velocity time and maximum velocity time refer to the time when the maximum blood flow velocity and the time when the minimum blood flow velocity are collected, respectively. These maximum and minimum velocity times can represent the delay characteristics of blood flow response. Therefore, the maximum and minimum velocity times are both considered as blood flow velocity features.

[0042] Specifically, the calculation of the brain response value based on the speed change rate group includes: The brain response value was calculated using the following formula: ; in, Represents the brain response value, Indicates the number of speed change rates in the speed change rate group, Indicates the speed change rate group The rate of change of speed, Indicates the speed change rate group The rate of change of speed, Indicates the sampling time interval.

[0043] Specifically, the blood flow feature extraction is performed on the blood flow data to obtain a blood flow feature group, including: Identify the test sampling period group, and extract the test sampling periods in the test sampling period group in sequence; determining a test light intensity curve corresponding to the test sampling period in the blood flow data, and converting the test light intensity curve into an electric field autocorrelation curve; Discretizing the electric field autocorrelation curve to obtain a fitting data point group, wherein the fitting data point group includes a plurality of fitting data points; The data is fitted using the fitting data point group to obtain the blood flow; Summarizing the blood flow corresponding to each experimental sampling period in the experimental sampling period group to obtain a blood flow sequence; Calculate the average blood flow of the blood flow series; Identifying a maximum blood flow rate and a minimum blood flow rate in a blood flow sequence, and determining a maximum flow time and a minimum flow time corresponding to the maximum blood flow rate and the minimum blood flow rate, respectively; The average blood flow, maximum blood flow, minimum blood flow, maximum flow time and minimum flow time are combined to obtain a blood flow feature group.

[0044] It is clear that the test sampling period group refers to a combination of multiple test sampling periods, and the test sampling period refers to the time period between the initial sampling moment and the intermediate sampling moment in the step of recording blood flow velocity data and blood flow data. The test light intensity curve refers to the light intensity autocorrelation curve collected during the test sampling period. The electric field autocorrelation curve refers to the electric field fluctuation correlation function converted from the light intensity autocorrelation curve through the Siegert relationship. The electric field autocorrelation curve can represent the time correlation of the photon electric field (its decay rate reflects the movement speed of red blood cells). Since the light intensity autophase curve contains detection noise and cannot be directly used for blood flow calculation, it is necessary to convert the test light intensity curve into the electric field autocorrelation curve, wherein the formula for converting the test light intensity curve into the electric field autocorrelation curve is: ,in, represents the electric field autocorrelation curve, represents the calibration coefficient, Represents the time delay variable, where the calibration coefficient can be selected as: 0.3-0.5. The time delay variable refers to the time difference between the arrival times of two photons in the autocorrelation function (unit: ns).

[0045] Furthermore, the fitting data point group refers to a combination of multiple discrete fitting data points, wherein the fitting data points refer to the discrete coordinate points in the electric field autocorrelation curve, and the fitting data points are expressed as: ,in, represents the fitted data points, Indicates the horizontal coordinate of the curve corresponding to the fitted data point in the electric field autocorrelation curve, Represents the vertical coordinate of the curve corresponding to the fitted data point in the electric field autocorrelation curve.

[0046] It should be explained that the blood flow sequence refers to a permutation and combination of multiple blood flow rates with time tags. The average blood flow refers to the average of multiple blood flow rates in the blood flow sequence. The maximum blood flow and minimum blood flow refer to the blood flow with the largest and smallest values ​​in the blood flow sequence, respectively. The maximum blood flow and minimum blood flow can reflect the maximum perfusion capacity (i.e., the maximum blood flow value) and the minimum ischemic threshold (i.e., the blood flow during ischemia) of the microvasculature during physiological testing tasks. The maximum flow time and minimum flow time refer to the time when the maximum blood flow and the time when the minimum blood flow are collected, respectively. The maximum flow time and minimum flow time can reflect the response speed and recovery speed of the microvasculature to stimulation. Therefore, the maximum blood flow, minimum blood flow, maximum flow time, and minimum flow time can all be used as blood flow characteristics.

[0047] In detail, performing data fitting using the fitting data point group to obtain blood flow includes: Constructing a multi-order blood flow fitting formula, wherein the multi-order blood flow fitting formula includes factors to be fitted, and the factors to be fitted include blood flow; Using a preset fitting algorithm and a multi-order blood flow fitting formula, data fitting is performed on the factor to be fitted to obtain a fitting factor; Based on the fitting factors, the blood flow is calculated.

[0048] Importantly, the multi-order blood flow fitting formula is expressed as: ; in, Represents the order of the multi-order blood flow fitting formula, which is set manually. Optionally, Set to 3, Indicates 2 to The constant between , s refers to the average transmission distance of photons in the optical detector, which can be determined by Monte Carlo (MC) simulation of photon migration, represents the factorial symbol, Represents the factor to be fitted, where the factor to be fitted refers to the unknown quantity that needs to be fitted. The factor to be fitted is expressed as: , where k represents the wave vector value of light in the medium (human tissue). The wave vector value is obtained by calculating based on the wavelength of light in human tissue and the optical properties of the tissue. The scattering coefficient of human tissue is obtained by measuring the degree of light scattering by human tissue and calculating it in combination with relevant formulas. The above wave vector value and scattering coefficient are obtained by existing technologies and will not be described in detail here. Indicates blood flow.

[0049] Furthermore, the fitting algorithm refers to a nonlinear least squares method. The fitting factor refers to a specific value of the factor to be fitted obtained after data fitting. Calculating blood flow based on the fitting factor refers to calculating blood flow using an expression for the fitting factor (where the expression for the fitting factor is the same as the expression for the factor to be fitted, and all parameters in the factor to be fitted, except blood flow, are known quantities).

[0050] S4. Summarize the test data to obtain a test data set, and label the test data set for clinical brain abnormalities to obtain a labeled data set.

[0051] It should be explained that for subsequent neural network training, the experimental data needs to be labeled.

[0052] Specifically, the step of labeling the clinical abnormalities of the brain in the test dataset to obtain a labeled dataset includes: Extracting test data sequentially from the test data set, determining a target patient corresponding to the test data, and obtaining a patient clinical indicator group of the target patient; Based on the patient clinical indicator group, performing brain disease analysis on the target patient to obtain a brain disease category group, wherein the brain disease category group includes one or more brain disease categories; Constructing a total disease category vector, wherein each vector element in the total disease category vector represents a brain disease; Use the brain disease category group to numerically label the total disease category vector to obtain the disease label vector; Supplementing the disease marker vector to the test data to obtain marker data; Aggregate the labeled data to obtain a labeled dataset.

[0053] It will be understood that the target patient refers to the test patient corresponding to the test data, and the patient clinical indicator group refers to a combination of multiple patient clinical indicators, wherein the patient clinical indicator refers to a numerical value that quantifies the target patient's brain disease, such as: cerebral infarction volume (mm³), cognitive score (MMSE), blood flow autoregulation index (ARI), etc. The brain disease category refers to the clinical manifestation of brain disease set by the physician, such as: cerebral infarction, Alzheimer's disease, vascular dementia, etc. The brain disease analysis of the target patient based on the patient clinical indicator group to obtain the brain disease category group means: the relevant physician determines the brain disease category of the target patient based on the patient clinical indicator group.

[0054] The total disease category vector points to a vector whose quantitative elements are all 0. The use of the brain disease category group to numerically mark the total disease category vector to obtain the disease marking vector refers to: marking the vector elements corresponding to the brain disease category group in the total disease category vector as the numerical value 1, and the marked total disease category vector is the disease marking vector.

[0055] For example, the total category vector of a disease is (vector element 1=0 vector element 2=0 vector element 3=0), where vector element 1, vector element 2 and vector element 3 represent disease category 1, disease category 2 and disease category 3 respectively. The brain disease category group of a target patient includes disease category 1 and disease category 2, then the vector elements corresponding to disease category 1 and disease category 2 are set to the value 1, and the disease label vector is: (1 1 0).

[0056] S5. Use the labeled dataset to train the pre-built neural network to obtain a brain disease analysis model.

[0057] It is clear that the training is supervised training, and the optional neural network models are: multi-layer perceptron (MLP), long short-term memory network (LSTM), convolutional neural network (CNN), etc., and the training process is a common step in neural network training, which will not be repeated here.

[0058] Furthermore, the brain disease analysis model refers to a trained neural network, and the output value of the model is a plurality of probability values, and the plurality of probability values ​​are respectively the probability values ​​of occurrence of each brain disease category.

[0059] S6. Receive a brain disease warning instruction, identify patients to be warned based on the brain disease warning instruction, measure cerebral blood flow on the patients to be warned, and obtain real-time cerebral blood flow data, wherein the real-time cerebral blood flow data includes: physiological characteristics of the warning patients, real-time blood flow velocity characteristics, and real-time blood flow characteristics.

[0060] It is understood that the brain disease warning instruction refers to a manually initiated instruction to issue a brain disease warning to a specific patient, and the patient to be warned refers to the specific patient specified in the brain disease warning instruction. The real-time cerebral blood flow data refers to the cerebral blood flow data of the patient to be warned, wherein the physiological characteristics of the warning patient, the real-time blood flow velocity characteristics, and the real-time blood flow characteristics refer to the patient's physiological characteristics, blood flow velocity characteristics, and blood flow characteristics in the real-time cerebral blood flow data, respectively. The method for obtaining the above-mentioned real-time cerebral blood flow data is the same as the method for obtaining the test data, and will not be repeated here.

[0061] S7. Perform brain disease analysis on the patient to be warned based on the real-time cerebral blood flow data and the brain disease analysis model to obtain a clinical abnormality feature group, wherein the clinical abnormality features in the clinical abnormality feature group are all probability values.

[0062] It can be understood that, based on the real-time cerebral blood flow data and the brain disease analysis model, brain disease analysis of the patient to be warned refers to: inputting the real-time cerebral blood flow data into the brain disease analysis model, and the model will output a set of data, which are: the probability of occurrence of the brain disease category corresponding to each vector element in the total disease category vector of the patient to be warned, that is, the clinical abnormality feature refers to the probability of a certain brain disease category occurring in the patient to be warned, and multiple clinical abnormality features constitute a clinical abnormality feature group.

[0063] S8. Generate an early warning analysis report based on the pre-built large language model and clinical abnormality feature group, and upload the early warning analysis report to the preset physician diagnosis and treatment terminal to complete the brain disease early warning based on the cerebral blood flow automatic regulation algorithm.

[0064] It should be clarified that the "large language model" refers to a generative AI model based on the Transformer architecture, such as GPT-4. Generating an early warning analysis report based on a pre-built large language model and a set of clinical abnormality features means inputting a disease probability vector into the large language model to generate a natural language text containing a risk description. The "early warning analysis report" refers to a natural language description of the risk of brain disease in a patient under early warning. For example, the early warning analysis report for a patient under early warning indicated a 72% risk of cerebral infarction (delayed blood flow response) and a 35% risk of Alzheimer's disease. Further MRI examination is recommended.

[0065] Furthermore, the physician diagnosis and treatment terminal refers to a cloud platform that provides patient data to physicians. After uploading the early warning analysis report to the physician diagnosis and treatment terminal, the relevant physician will receive the early warning analysis report and make further judgments based on the early warning analysis report.

[0066] In order to solve the problems described in the background technology, the present invention first constructs a cerebral blood flow detection unit, which realizes comprehensive monitoring of cerebral blood flow velocity and flow, and provides a hardware foundation for subsequent acquisition of accurate cerebral blood flow characteristic data. A physiological test task group is set, and based on the physiological test task group and the test unit, cerebral blood flow detection is performed to obtain test data. This step simulates the changes in cerebral blood flow under different physiological states by setting a variety of physiological test tasks, and can more comprehensively collect cerebral blood flow characteristic data of test patients under various conditions, including physiological characteristics, blood flow velocity characteristics and blood flow characteristics, enriching the data dimension, and providing more sufficient and diverse data support for the subsequent establishment of an accurate brain disease analysis model, which helps to improve the model's ability to recognize different brain disease characteristics. Then, the test data set is labeled for brain clinical abnormalities to obtain a labeled data set. This step associates brain clinical abnormalities with corresponding data, providing a clearly labeled data set for subsequent neural network training. The training data set enables the neural network to learn the mapping relationship between cerebral blood flow characteristics and brain diseases. Then, by using the labeled data set to perform supervised training on the neural network, the model can learn the association between cerebral blood flow characteristics and brain disease categories, thereby constructing a model with brain disease identification and analysis capabilities, providing an intelligent analysis tool for subsequent brain disease risk assessment of patients to be warned. Furthermore, by inputting real-time cerebral blood flow data into the brain disease analysis model, the risk probability of various brain diseases in patients to be warned can be quickly and accurately analyzed, and the clinical abnormality feature group is intuitively presented in the form of probability values, providing a quantitative risk assessment for clinicians. Finally, the large language model is used to convert the complex clinical abnormality feature probability value into an easy-to-understand warning analysis report, and it is promptly transmitted to the physician, achieving efficient information transmission and visual presentation, facilitating the physician to quickly understand the patient's brain disease risk situation and assisting the physician in making decisions. Therefore, the present invention can improve the intelligence level of brain warning and enhance the accuracy of brain warning.

[0067] like Figure 2 2 is a functional module diagram of a brain disease early warning system based on a cerebral blood flow autoregulation algorithm provided by an embodiment of the present invention.

[0068] The brain disease early warning system 100 based on the cerebral blood flow autoregulation algorithm of the present invention can be installed in an electronic device. Depending on the functions implemented, the brain disease early warning system 100 based on the cerebral blood flow autoregulation algorithm can include a test unit construction module 101, a test data marking module 102, a patient data collection module 103, and a warning report generation module 104. The modules described in the present invention, also referred to as units, refer to a series of computer program segments that can be executed by an electronic device processor and can perform a fixed function, which are stored in the memory of the electronic device.

[0069] The test unit construction module 101 is used to identify a test patient set and construct a cerebral blood flow detection unit, wherein the cerebral blood flow detection unit includes: a transcranial Doppler detector, an optical detector, a blood flow velocity analyzer, and a blood flow analyzer, and the blood flow velocity analyzer is connected to the transcranial Doppler detector, and the blood flow analyzer is connected to the optical detector. Test patients are sequentially extracted from the test patient set, and the test unit is constructed based on the test patients and the cerebral blood flow detection unit; The test data labeling module 102 is used to set a physiological test task group, and perform cerebral blood flow detection based on the physiological test task group and the test unit to obtain test data, wherein the physiological test task group includes one or more physiological test tasks, and the test data includes: physiological characteristics, blood flow velocity characteristics, and blood flow characteristics of the test patient; summarize the test data to obtain a test data set; and label the test data set for brain clinical abnormalities to obtain a labeled data set; The patient data acquisition module 103 is configured to train a pre-built neural network using a labeled data set to obtain a brain disease analysis model, receive a brain disease warning instruction, identify patients to be warned based on the brain disease warning instruction, and measure cerebral blood flow in the patients to be warned to obtain real-time cerebral blood flow data, wherein the real-time cerebral blood flow data includes: physiological characteristics of the patients to be warned, real-time blood flow velocity characteristics, and real-time blood flow characteristics; The warning report generation module 104 is used to perform brain disease analysis on the patient to be warned based on real-time cerebral blood flow data and a brain disease analysis model to obtain a clinical abnormality feature group, wherein the clinical abnormal features in the clinical abnormality feature group are all probability values, generate a warning analysis report based on a pre-built large language model and the clinical abnormality feature group, and upload the warning analysis report to a preset physician diagnosis and treatment terminal.

[0070] In detail, each module in the brain disease early warning system 100 based on the cerebral blood flow automatic regulation algorithm in the embodiment of the present invention adopts the same Figure 1 The technical means are the same as the brain disease early warning method based on the cerebral blood flow automatic regulation algorithm described in, and can produce the same technical effects, so I will not go into details here.

[0071] like Figure 3 FIG. 1 is a schematic diagram of the structure of an electronic device for implementing a brain disease early warning method based on a cerebral blood flow autoregulation algorithm according to an embodiment of the present invention.

[0072] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a brain disease early warning method program based on a cerebral blood flow autoregulation algorithm.

[0073] The memory 11 includes at least one type of readable storage medium, including flash memory, a mobile hard drive, a multimedia card, a card-type memory (e.g., SD or DX memory), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 11 may be an internal storage unit of the electronic device 1, such as a mobile hard drive of the electronic device 1. In other embodiments, the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in mobile hard drive, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. equipped on the electronic device 1. Furthermore, the memory 11 includes both internal storage units of the electronic device 1 and external storage devices. The memory 11 can be used not only to store application software installed in the electronic device 1 and various types of data, such as the code of a brain disease early warning method program based on a cerebral blood flow autoregulation algorithm, but also to temporarily store data that has been output or is about to be output.

[0074] In some embodiments, the processor 10 may be comprised of an integrated circuit, such as a single packaged integrated circuit or a combination of multiple packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control core (control unit) of the electronic device, connecting the various components of the electronic device using various interfaces and circuits. It executes programs or modules stored in the memory 11 (e.g., a brain disease early warning method program based on a cerebral blood flow autoregulation algorithm) and accesses data stored in the memory 11 to perform various functions and process data.

[0075] The bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus 12 may be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to enable communication between the memory 11 and at least one processor 10, etc.

[0076] Figure 3 Only the electronic device with components is shown, and it can be understood by those skilled in the art that Figure 3The structure shown does not constitute a limitation on the electronic device 1 , and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0077] For example, although not shown, the electronic device 1 may further include a power source (e.g., a battery) to power various components. Preferably, the power source may be logically connected to the at least one processor 10 via a power management system, thereby enabling functions such as charge management, discharge management, and power consumption management through the power management system. The power source may further include any components such as one or more DC or AC power sources, a recharging system, a power failure detection circuit, a power converter or inverter, and a power status indicator. The electronic device 1 may further include various sensors, Bluetooth modules, Wi-Fi modules, etc., which are not further described here.

[0078] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0079] Optionally, the electronic device 1 may further include a user interface, which may be a display or an input unit (such as a keyboard). Optionally, the user interface may also be a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. The display may also be appropriately referred to as a display screen or a display unit, and is used to display information processed by the electronic device 1 and to display a visual user interface.

[0080] The brain disease early warning method program based on the cerebral blood flow automatic regulation algorithm stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can achieve the following: Identify the test patient set and construct a cerebral blood flow detection unit, wherein the cerebral blood flow detection unit includes: a transcranial Doppler detector, an optical detector, a blood flow velocity analyzer, and a blood flow analyzer, wherein the blood flow velocity analyzer is connected to the transcranial Doppler detector, and the blood flow analyzer is connected to the optical detector; Extract test patients in sequence from the test patient set, and construct test units based on the test patients and cerebral blood flow detection units; A physiological test task group is set up, and based on the physiological test task group and the test unit, cerebral blood flow detection is performed to obtain test data, wherein the physiological test task group includes one or more physiological test tasks, and the test data includes: physiological characteristics of the test patient, blood flow velocity characteristics, and blood flow characteristics; Aggregating the test data to obtain a test data set, and labeling the test data set for brain clinical abnormalities to obtain a labeled data set; Use labeled datasets to train pre-built neural networks to obtain brain disease analysis models; receiving a brain disease warning instruction, identifying a patient to be warned based on the brain disease warning instruction, measuring cerebral blood flow of the patient to be warned, and obtaining real-time cerebral blood flow data, wherein the real-time cerebral blood flow data includes: physiological characteristics of the patient to be warned, real-time blood flow velocity characteristics, and real-time blood flow characteristics; Based on the real-time cerebral blood flow data and brain disease analysis model, brain disease analysis is performed on the patients to be warned, and a clinical abnormality feature group is obtained, wherein the clinical abnormality features in the clinical abnormality feature group are all probability values; An early warning analysis report is generated based on the pre-built large language model and clinical abnormality feature group, and the early warning analysis report is uploaded to the preset physician diagnosis and treatment terminal to complete the brain disease early warning based on the cerebral blood flow automatic regulation algorithm.

[0081] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to Figures 1 to 3 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0082] Furthermore, if the modules / units integrated into the electronic device 1 are implemented as software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium. The computer-readable storage medium may be volatile or non-volatile. For example, the computer-readable medium may include any entity or system capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0083] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program. When the computer program is executed by a processor of an electronic device, the computer program can implement: Identify the test patient set and construct a cerebral blood flow detection unit, wherein the cerebral blood flow detection unit includes: a transcranial Doppler detector, an optical detector, a blood flow velocity analyzer, and a blood flow analyzer, wherein the blood flow velocity analyzer is connected to the transcranial Doppler detector, and the blood flow analyzer is connected to the optical detector; Extract test patients in sequence from the test patient set, and construct test units based on the test patients and cerebral blood flow detection units; A physiological test task group is set up, and based on the physiological test task group and the test unit, cerebral blood flow detection is performed to obtain test data, wherein the physiological test task group includes one or more physiological test tasks, and the test data includes: physiological characteristics of the test patient, blood flow velocity characteristics, and blood flow characteristics; Aggregating the test data to obtain a test data set, and labeling the test data set for brain clinical abnormalities to obtain a labeled data set; Use labeled datasets to train pre-built neural networks to obtain brain disease analysis models; receiving a brain disease warning instruction, identifying a patient to be warned based on the brain disease warning instruction, measuring cerebral blood flow of the patient to be warned, and obtaining real-time cerebral blood flow data, wherein the real-time cerebral blood flow data includes: physiological characteristics of the patient to be warned, real-time blood flow velocity characteristics, and real-time blood flow characteristics; Based on the real-time cerebral blood flow data and brain disease analysis model, brain disease analysis is performed on the patients to be warned, and a clinical abnormality feature group is obtained, wherein the clinical abnormality features in the clinical abnormality feature group are all probability values; An early warning analysis report is generated based on the pre-built large language model and clinical abnormality feature group, and the early warning analysis report is uploaded to the preset physician diagnosis and treatment terminal to complete the brain disease early warning based on the cerebral blood flow automatic regulation algorithm.

[0084] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only exemplary, and actual implementations may have other division methods.

[0085] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network elements. Some or all of the modules may be selected to achieve the purpose of the solution of this embodiment according to actual needs.

[0086] In addition, the functional modules in various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or hardware plus software functional modules.

[0087] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A brain disease early warning method based on cerebral blood flow automatic regulation algorithm, characterized in that: The method comprises: Identify the test patient set and construct a cerebral blood flow detection unit, wherein the cerebral blood flow detection unit includes: a transcranial Doppler detector, an optical detector, a blood flow velocity analyzer, and a blood flow analyzer, wherein the blood flow velocity analyzer is connected to the transcranial Doppler detector, and the blood flow analyzer is connected to the optical detector; Extract test patients in sequence from the test patient set, and construct test units based on the test patients and cerebral blood flow detection units; A physiological test task group is set up, and based on the physiological test task group and the test unit, cerebral blood flow detection is performed to obtain test data, wherein the physiological test task group includes one or more physiological test tasks, and the test data includes: physiological characteristics of the test patient, blood flow velocity characteristics, and blood flow characteristics; Aggregating the test data to obtain a test data set, and labeling the test data set for brain clinical abnormalities to obtain a labeled data set; Use labeled datasets to train pre-built neural networks to obtain brain disease analysis models; receiving a brain disease warning instruction, identifying a patient to be warned based on the brain disease warning instruction, measuring cerebral blood flow of the patient to be warned, and obtaining real-time cerebral blood flow data, wherein the real-time cerebral blood flow data includes: physiological characteristics of the patient to be warned, real-time blood flow velocity characteristics, and real-time blood flow characteristics; Based on the real-time cerebral blood flow data and brain disease analysis model, brain disease analysis is performed on the patients to be warned, and a clinical abnormality feature group is obtained, wherein the clinical abnormality features in the clinical abnormality feature group are all probability values; An early warning analysis report is generated based on the pre-built large language model and clinical abnormality feature group, and the early warning analysis report is uploaded to the preset physician diagnosis and treatment terminal to complete the brain disease early warning based on the cerebral blood flow automatic regulation algorithm.

2. The brain disease early warning method based on cerebral blood flow autoregulation algorithm according to claim 1, characterized in that: The test unit is constructed according to the test patient and the cerebral blood flow detection unit, including: The test patient is fixed using a cerebral blood flow detection unit to obtain a test unit, wherein the test unit includes: a test patient and a cerebral blood flow detection unit, the transcranial Doppler detector in the cerebral blood flow detection unit is fixed on the temporal window bone of the test patient, and the transcranial Doppler detector is connected to the blood flow velocity analyzer via a connecting line, the optical detector is fixed 2 cm above the eyebrow of the test patient, and the optical detector is connected to the blood flow analyzer via a connecting line, and an optical probe and a light source are pre-installed in the optical detector.

3. The brain disease early warning method based on cerebral blood flow autoregulation algorithm according to claim 2, characterized in that: The physiological test task group and the test unit are used to perform cerebral blood flow detection and obtain test data, including: extracting a first test task from the physiological test task group; Based on the test patient in the test unit, perform a first test task, and record blood flow velocity data and blood flow volume data in the step of performing the first test task; Extracting blood flow velocity features from the blood flow velocity data to obtain a blood flow velocity feature group; Extracting blood flow features from the blood flow data to obtain a blood flow feature group; Acquiring a test patient physiological characteristic group of the test patient, wherein the test patient physiological characteristic group includes a plurality of test patient physiological characteristics; The physiological characteristic group, blood velocity characteristic group and blood flow characteristic group of the test patients are combined to obtain a single-task characteristic group; The single-task feature groups are aggregated to obtain multiple single-task feature groups, and the multiple single-task feature groups are merged to obtain test data.

4. The brain disease early warning method based on cerebral blood flow autoregulation algorithm according to claim 3, characterized in that: The step of recording the blood flow velocity data and blood flow volume data in the step of performing the first test task includes: Determining an initial sampling time in the step of performing the first test task; Calculate the intermediate sampling time based on the initial sampling time and the preset sampling time interval; The light intensity autocorrelation curve between the initial sampling moment and the intermediate sampling moment is recorded by an optical detector, and the blood flow velocity is detected by a transcranial Doppler detector; Recording the intermediate sampling moment as the initial sampling moment, and returning to the step of calculating the intermediate sampling moment based on the initial sampling moment and the preset sampling time interval, until the intermediate sampling moment is not less than the preset end sampling moment; The blood flow velocity and light intensity autocorrelation curves are respectively summarized to obtain a blood flow velocity group and a light intensity autocorrelation curve group, and the blood flow velocity group and the light intensity autocorrelation curve group are respectively recorded as blood flow velocity data and blood flow volume data.

5. The brain disease early warning method based on cerebral blood flow autoregulation algorithm according to claim 4, characterized in that: The blood flow velocity feature extraction is performed on the blood flow velocity data to obtain a blood flow velocity feature group, including: Obtain the resting blood flow rate group of the trial patients; According to the preset response time, the blood flow velocity data is divided into instantaneous blood flow velocity groups, the instantaneous blood flow velocities are sequentially extracted from the instantaneous blood flow velocity groups, and the average blood flow velocity of the static blood flow velocity groups is calculated; The velocity change rate is calculated based on the average blood flow velocity and the instantaneous blood flow velocity, where the velocity change rate is expressed as: ； in, represents the rate of change of speed, represents the instantaneous blood flow velocity, represents the mean blood flow velocity; Summarizing the velocity change rates corresponding to each instantaneous blood flow velocity in the instantaneous blood flow velocity group to obtain a velocity change rate group, and calculating the brain response value based on the velocity change rate group; Confirming the maximum blood flow velocity and the minimum blood flow velocity in the bleeding flow velocity data, and determining the minimum velocity time and the maximum velocity time corresponding to the maximum blood flow velocity and the minimum blood flow velocity, respectively; The brain response value, maximum blood flow velocity, minimum blood flow velocity, minimum velocity time and maximum velocity time are combined to obtain a blood flow velocity feature group.

6. The brain disease early warning method based on cerebral blood flow autoregulation algorithm according to claim 5, characterized in that: The calculating of the brain response value based on the speed change rate group includes: The brain response value was calculated using the following formula: ； in, Represents the brain response value, Indicates the number of speed change rates in the speed change rate group, Indicates the speed change rate group The rate of change of speed, Indicates the speed change rate group The rate of change of speed, Indicates the sampling time interval.

7. The brain disease early warning method based on cerebral blood flow autoregulation algorithm according to claim 6, characterized in that: The blood flow feature extraction is performed on the blood flow data to obtain a blood flow feature group, including: Identify the test sampling period group, and extract the test sampling periods in the test sampling period group in sequence; determining a test light intensity curve corresponding to the test sampling period in the blood flow data, and converting the test light intensity curve into an electric field autocorrelation curve; Discretizing the electric field autocorrelation curve to obtain a fitting data point group, wherein the fitting data point group includes a plurality of fitting data points; The data is fitted using the fitting data point group to obtain the blood flow; Summarizing the blood flow corresponding to each experimental sampling period in the experimental sampling period group to obtain a blood flow sequence; Calculate the average blood flow of the blood flow series; Identifying a maximum blood flow rate and a minimum blood flow rate in a blood flow sequence, and determining a maximum flow time and a minimum flow time corresponding to the maximum blood flow rate and the minimum blood flow rate, respectively; The average blood flow, maximum blood flow, minimum blood flow, maximum flow time and minimum flow time are combined to obtain a blood flow feature group.

8. The brain disease early warning method based on cerebral blood flow autoregulation algorithm according to claim 7, characterized in that: The method of performing data fitting using the fitting data point group to obtain the blood flow rate includes: Constructing a multi-order blood flow fitting formula, wherein the multi-order blood flow fitting formula includes factors to be fitted, and the factors to be fitted include blood flow; Using a preset fitting algorithm and a multi-order blood flow fitting formula, data fitting is performed on the factor to be fitted to obtain a fitting factor; Based on the fitting factors, the blood flow is calculated.

9. The brain disease early warning method based on cerebral blood flow autoregulation algorithm according to claim 8, characterized in that: The step of labeling the clinical abnormalities of the brain on the test dataset to obtain a labeled dataset includes: Extracting test data sequentially from the test data set, determining a target patient corresponding to the test data, and obtaining a patient clinical indicator group of the target patient; performing brain disease analysis on the target patient based on the patient clinical indicator group to obtain a brain disease category group, wherein the brain disease category group includes one or more brain disease categories; Constructing a total disease category vector, wherein each vector element in the total disease category vector represents a brain disease; Use the brain disease category group to numerically label the total disease category vector to obtain the disease label vector; Supplementing the disease marker vector to the test data to obtain marker data; Aggregate the labeled data to obtain a labeled dataset.

10. A brain disease early warning system based on cerebral blood flow automatic regulation algorithm, characterized in that: The system comprises: A test unit construction module is used to identify a test patient set and construct a cerebral blood flow detection unit, wherein the cerebral blood flow detection unit includes: a transcranial Doppler detector, an optical detector, a blood flow velocity analyzer, and a blood flow analyzer, and the blood flow velocity analyzer is connected to the transcranial Doppler detector, and the blood flow analyzer is connected to the optical detector. Test patients are sequentially extracted from the test patient set, and the test unit is constructed based on the test patients and the cerebral blood flow detection unit; a test data labeling module, configured to set a physiological test task group, and perform cerebral blood flow detection based on the physiological test task group and the test unit to obtain test data, wherein the physiological test task group includes one or more physiological test tasks, and the test data includes: physiological characteristics, blood flow velocity characteristics, and blood flow characteristics of the test patient; summarize the test data to obtain a test data set; and label the test data set for brain clinical abnormalities to obtain a labeled data set; The patient data acquisition module is used to train a pre-built neural network using a labeled data set to obtain a brain disease analysis model, receive brain disease warning instructions, identify patients to be warned based on the brain disease warning instructions, measure cerebral blood flow in the patients to be warned, and obtain real-time cerebral blood flow data, wherein the real-time cerebral blood flow data includes: physiological characteristics of the patients to be warned, real-time blood flow velocity characteristics, and real-time blood flow characteristics; The early warning report generation module is used to analyze brain diseases of patients under warning based on real-time cerebral blood flow data and brain disease analysis models, and obtain a clinical abnormality feature group, where the clinical abnormal features in the clinical abnormality feature group are all probability values. The early warning analysis report is generated based on the pre-built large language model and the clinical abnormality feature group, and the early warning analysis report is uploaded to the preset physician diagnosis and treatment terminal.

Images